CGN_taiyuan_104/data/ClassIdentifier.py

import datetime
import traceback

import numpy as np
from pandas import DataFrame

from utils.log.trans_log import logger


class ClassIdentifier(object):
    """
    分类标识 -1:停机 0:好点  1:欠发功率点；2:超发功率点；3:额定风速以上的超发功率点 4: 限电
    """

    def __init__(self, wind_turbine_number="", origin_df: DataFrame = None,
                 wind_velocity='wind_velocity',
                 active_power='active_power',
                 pitch_angle_blade='pitch_angle_blade_1',
                 rated_power=1500, cut_out_speed=20):
        """
        :param origin_df: The pandas DataFrame containing the input data.
        :param wind_velocity: 风速字段
        :param active_power: 有功功率字段
        :param pitch_angle_blade: 桨距角
        :param rated_power: 额定功率
        :param cut_out_speed: 切出风速
        """
        self.wind_turbine_number = wind_turbine_number
        self.wind_velocity = wind_velocity
        self.active_power = active_power
        self.pitch_angle_blade = pitch_angle_blade
        self.rated_power = rated_power  # 额定功率1500kw,可改为2000kw
        self.cut_out_speed = cut_out_speed

        if self.rated_power is None:
            logger.info(f"{wind_turbine_number} WARNING:rated_power配置为空的")
            self.rated_power = 1500

        if self.cut_out_speed is None:
            logger.info(f"{cut_out_speed} WARNING:cut_out_speed配置为空的")
            self.cut_out_speed = 20

        self.df = origin_df

    def identifier(self):
        # 风速 和 有功功率 df
        # self.df = self.df[[self.wind_velocity, self.active_power, "pitch_angle_blade_1"]]
        self.df.reset_index(inplace=True)

        if len(self.df[self.active_power].unique()) <= 10:
            self.df['lab'] = -1
            return self.df

        wind_and_power_df_count = self.df.shape[0]
        power_max = self.df[self.active_power].max()
        power_rated = np.ceil(power_max / 100) * 100
        v_cut_out = self.cut_out_speed
        # 网格法确定风速风向分区数量，功率方向分区数量，
        power_bin_count = int(np.ceil(power_rated / 25))  # 功率分区间隔25kW
        velocity_bin_count = int(np.ceil(v_cut_out / 0.25))  # 风速分区间隔0.25m/s

        # 存储功率大于零的运行数据
        power_gt_zero_array = np.zeros([wind_and_power_df_count, 2], dtype=float)
        power_gt_zero_array_count = 0
        for i in range(wind_and_power_df_count):
            if self.df.loc[i, self.active_power] > 0:
                power_gt_zero_array[power_gt_zero_array_count, 0] = self.df.loc[i, self.wind_velocity]
                power_gt_zero_array[power_gt_zero_array_count, 1] = self.df.loc[i, self.active_power]

                power_gt_zero_array_count = power_gt_zero_array_count + 1

        # 统计各网格落入的散点个数
        x_box_number = np.zeros([power_bin_count, velocity_bin_count], dtype=int)

        n_which_p = -1
        n_which_v = -1
        for i in range(power_gt_zero_array_count):
            for m in range(power_bin_count):
                if m * 25 < power_gt_zero_array[i, 1] <= (m + 1) * 25:
                    n_which_p = m
                    break

            for n in range(velocity_bin_count):
                if (n * 0.25 + 0.125) < power_gt_zero_array[i, 0] <= ((n + 1) * 0.25 + 0.125):
                    n_which_v = n
                    break

            if n_which_p > -1 and n_which_v > -1:
                x_box_number[n_which_p, n_which_v] = x_box_number[n_which_p, n_which_v] + 1

        # 在功率方向将网格内散点绝对个数转换为相对百分比，备用
        power_box_percent = np.zeros([power_bin_count, velocity_bin_count], dtype=float)

        # 功率方向统计
        power_bin_sum = np.zeros(power_bin_count, dtype=int)

        for i in range(power_bin_count):
            power_bin_sum[i] = sum(x_box_number[i, :])
            # for m in range(velocity_bin_count):
            #     power_bin_sum[i] = power_bin_sum[i] + x_box_number[i, m]

            for m in range(velocity_bin_count):
                if power_bin_sum[i] > 0:
                    power_box_percent[i, m] = x_box_number[i, m] / power_bin_sum[i] * 100

        # 在风速方向将网格内散点绝对个数转换为相对百分比，备用
        v_box_percent = np.zeros([power_bin_count, velocity_bin_count], dtype=float)
        v_bin_sum = np.zeros(velocity_bin_count, dtype=int)

        for i in range(velocity_bin_count):
            v_bin_sum[i] = sum(x_box_number[:, i])
            # for m in range(power_bin_count):
            #     v_bin_sum[i] = v_bin_sum[i] + x_box_number[m, i]

            for m in range(power_bin_count):
                if v_bin_sum[i] > 0:
                    v_box_percent[m, i] = x_box_number[m, i] / v_bin_sum[i] * 100

        # 以水平功率带方向为准，分析每个水平功率带中，功率主带中心，即找百分比最大的网格位置。
        p_box_max_index = np.zeros(power_bin_count, dtype=int)  # 水平功率带最大网格位置索引
        p_box_max_p = np.zeros(power_bin_count, dtype=int)  # 水平功率带最大网格百分比

        for m in range(power_bin_count):
            # 确定每一水平功率带的最大网格位置索引即百分比值
            p_box_max_p[m], p_box_max_index[m] = power_box_percent[m, :].max(), power_box_percent[m, :].argmax()

        # 切入风速特殊处理，如果切入风速过于偏右，向左拉回
        # todo 为什么第一行数据的索引值 > 14个就要往左拉回,还有是不是不叫切入风速,这个是 落入这个区间功率最多的个数的索引值
        if p_box_max_index[0] > 14:
            p_box_max_index[0] = 9

        # 以水平功率带方向为基准，进行分析
        dot_dense_left_right = np.zeros([power_bin_count, 2], dtype=int)  # 存储每一水平功率带的功率主带以最大网格为中心，向向左，向右扩展的网格数
        dot_valve = 90  # 从中心向左右对称扩展网格的散点百分比和的阈值。

        for i in range(power_bin_count - 6):  # 从最下层水平功率带1开始，向上到第PNum-6个水平功率带（额定功率一下水平功率带），逐一分析
            p_dot_dense_sum = p_box_max_p[i]  # 以中心最大水平功率带为基准，向左向右对称扩展网格，累加各网格散点百分比
            i_spread_right = 1
            i_spread_left = 1
            while p_dot_dense_sum < dot_valve:

                if (p_box_max_index[i] + i_spread_right) < velocity_bin_count - 1:
                    # 向右侧扩展
                    p_dot_dense_sum = p_dot_dense_sum + power_box_percent[i, p_box_max_index[i] + i_spread_right]
                    i_spread_right = i_spread_right + 1

                if (p_box_max_index[i] + i_spread_right) > velocity_bin_count - 1:
                    break

                if (p_box_max_index[i] - i_spread_left) > 0:
                    # 向左侧扩展
                    p_dot_dense_sum = p_dot_dense_sum + power_box_percent[i, p_box_max_index[i] - i_spread_left]
                    i_spread_left = i_spread_left + 1

                if (p_box_max_index[i] - i_spread_left) <= 0:
                    break

            i_spread_right = i_spread_right - 1
            i_spread_left = i_spread_left - 1
            # 向左右对称扩展完毕

            dot_dense_left_right[i, 0] = i_spread_left
            dot_dense_left_right[i, 1] = i_spread_right

        main_band_right = np.median(dot_dense_left_right[:, 1])

        # 散点向右显著延展分布的水平功率带为限功率水平带
        # 各水平功率带是否为限功率标识，==1：是；==0：不是
        power_limit = np.zeros([power_bin_count, 1], dtype=int)
        width_average = 0  # 功率主带平均宽度

        # todo 限功率主带判别阈值为什么要加3
        power_limit_valve = np.ceil(main_band_right) + 3  # 限功率主带判别阈值
        n_counter = 0

        for i in range(power_bin_count - 6):
            # 如果向右扩展网格数大于阈值，且该水平功率带点总数>20，是限功率
            if dot_dense_left_right[i, 1] > power_limit_valve and power_bin_sum[i] > 20:
                power_limit[i] = 1

            if dot_dense_left_right[i, 1] <= power_limit_valve:
                # 统计正常水平功率带右侧宽度
                width_average = width_average + dot_dense_left_right[i, 1]
                n_counter = n_counter + 1

        width_average = width_average / n_counter  # 功率主带平均宽度

        # 对限负荷水平功率带的最大网格较下面相邻层显著偏右，拉回
        for i in range(1, power_bin_count - 6):
            if power_limit[i] == 1 and abs(p_box_max_index[i] - p_box_max_index[i - 1]) > 5:
                p_box_max_index[i] = p_box_max_index[i - 1] + 1

        # 功率主带的右边界
        curve_width = int(np.ceil(width_average) + 2)

        # 数据异常需要剔除的网格标识，标识1：功率主带右侧的欠发网格；2：功率主带左侧的超发网格 3:额定功率以上的超发点
        b_box_remove = np.zeros([power_bin_count, velocity_bin_count], dtype=int)

        for m in range(power_bin_count - 6):
            for n in range(p_box_max_index[m] + curve_width, velocity_bin_count):
                b_box_remove[m, n] = 1

            for n in range(p_box_max_index[m] - curve_width, -1, -1):
                b_box_remove[m, n] = 2

        # 确定功率主带的左上拐点，即额定风速位置的网格索引
        curve_top = np.zeros(2, dtype=int)
        curve_top_valve = 3  # 网格的百分比阈值
        b_top_find = False
        for m in range(power_bin_count - 5, -1, -1):
            for n in range(velocity_bin_count):
                # 如左上角网格的百分比和散点个数大于阈值。
                if v_box_percent[m, n] > curve_top_valve and x_box_number[m, n] >= 10:
                    curve_top[0] = m
                    curve_top[1] = n
                    b_top_find = True
                    break

            if b_top_find:
                break

        isolate_valve = 3
        for m in range(power_bin_count - 6):
            for n in range(p_box_max_index[m] + curve_width, velocity_bin_count):
                if power_box_percent[m, n] < isolate_valve:
                    b_box_remove[m, n] = 1

        # 功率主带顶部宽度
        curve_width_t = 2
        for m in range(power_bin_count - curve_width_t - 1, power_bin_count):
            for n in range(velocity_bin_count):
                b_box_remove[m, n] = 3  # 网格为额定功率以上的超发点

        # 功率主带拐点左侧的欠发网格标识
        for m in range(power_bin_count - 5 - 1, power_bin_count):
            for n in range(curve_top[1] - 1):
                b_box_remove[m, n] = 2

        # 以网格的标识，决定该网格内数据的标识。dzwind_and_power_sel。散点在哪个网格，此网格的标识即为该点的标识
        # -1:停机 0:好点  1:欠发功率点；2:超发功率点；3:额定风速以上的超发功率点 4: 限电
        dzwind_and_power_sel = np.zeros(power_gt_zero_array_count, dtype=int)
        n_which_p = -1
        n_which_v = -1

        for i in range(power_gt_zero_array_count):
            for m in range(power_bin_count):
                if m * 25 < power_gt_zero_array[i, 1] <= (m + 1) * 25:
                    n_which_p = m
                    break

            for n in range(velocity_bin_count):
                if (n * 0.25 + 0.125) < power_gt_zero_array[i, 0] <= ((n + 1) * 0.25 + 0.125):
                    n_which_v = n
                    break

            if n_which_p > -1 and n_which_v > -1:
                if b_box_remove[n_which_p, n_which_v] == 1:
                    dzwind_and_power_sel[i] = 1

                if b_box_remove[n_which_p, n_which_v] == 2:
                    dzwind_and_power_sel[i] = 2

                if b_box_remove[n_which_p, n_which_v] == 3:
                    dzwind_and_power_sel[i] = 0  # 3  # 额定风速以上的超发功率点认为是正常点，不再标识。

        # 限负荷数据标识方法2：把数据切割为若干个窗口。对每一窗口，以第一个点为基准，连续nWindowLength个数据的功率在方差范围内，呈现显著水平分布的点
        n_window_length = 3
        limit_window = np.zeros(n_window_length, dtype=float)
        power_std = 15  # 功率波动方差
        n_window_num = int(np.floor(power_gt_zero_array_count / n_window_length))
        power_limit_up = self.rated_power - 300
        power_limit_low = 200
        for i in range(n_window_num):
            for j in range(n_window_length):
                limit_window[j] = power_gt_zero_array[i * n_window_length + j, 1]

            b_all_in_areas = 1
            for j in range(n_window_length):
                if limit_window[j] < power_limit_low or limit_window[j] > power_limit_up:
                    b_all_in_areas = 0

            if b_all_in_areas == 0:
                continue

            up_limit = limit_window[0] + power_std
            low_limit = limit_window[0] - power_std
            b_all_in_up_low = 1
            for j in range(1, n_window_length):
                if limit_window[j] < low_limit or limit_window[j] > up_limit:
                    b_all_in_up_low = 0

            if b_all_in_up_low == 1:
                for j in range(n_window_length):
                    dzwind_and_power_sel[i * n_window_length + j] = 4  # 标识窗口内的数据为限负荷数据

        for i in range(power_bin_count - 6):
            pv_left_down = np.zeros(2, dtype=float)
            pv_right_up = np.zeros(2, dtype=float)

            if (p_box_max_index[i + 1] - p_box_max_index[i]) >= 1:
                pv_left_down[0] = (p_box_max_index[i] + curve_width) * 0.25 + 0.125
                pv_left_down[1] = i * 25

                pv_right_up[0] = (p_box_max_index[i + 1] + curve_width) * 0.25 + 0.125
                pv_right_up[1] = (i + 1) * 25

                for m in range(power_gt_zero_array_count):
                    if pv_left_down[0] < power_gt_zero_array[m, 0] < pv_right_up[0] and \
                            pv_left_down[1] < power_gt_zero_array[m, 1] < pv_right_up[1]:  # 在该锯齿中
                        if (power_gt_zero_array[m, 1] - pv_left_down[1]) / (
                                power_gt_zero_array[m, 0] - pv_left_down[0]) > (
                                pv_right_up[1] - pv_left_down[1]) / (
                                pv_right_up[0] - pv_left_down[0]):  # 斜率大于对角连线，则在锯齿左上三角形中，选中
                            dzwind_and_power_sel[m] = 0

        self.df.loc[:, 'lab'] = -1
        self.df.loc[
            self.df[self.df[self.active_power] > 0].index, 'lab'] = dzwind_and_power_sel

        # 把部分欠发的优化为限电
        # 构建条件表达式
        cond1 = (self.df['lab'] == 1) & (
                (self.df[self.active_power] < self.rated_power * 0.75) &
                (self.df[self.pitch_angle_blade] > 0.5)
        )
        cond2 = (self.df['lab'] == 1) & (
                (self.df[self.active_power] < self.rated_power * 0.85) &
                (self.df[self.pitch_angle_blade] > 1.5)
        )
        cond3 = (self.df['lab'] == 1) & (
                (self.df[self.active_power] < self.rated_power * 0.9) &
                (self.df[self.pitch_angle_blade] > 2.5)
        )

        # 使用逻辑或操作符|合并条件
        combined_condition = cond1 | cond2 | cond3
        self.df.loc[combined_condition, 'lab'] = 4

        self.df.loc[self.df[self.active_power] <= 0, 'lab'] = -1

        self.df.reset_index(drop=True, inplace=True)
        if 'index' in self.df.columns:
            del self.df['index']
        return self.df

    def run(self):
        # Implement your class identification logic here
        begin = datetime.datetime.now()
        logger.info(f"打标签开始,风机号:{self.wind_turbine_number},数量:{self.df.shape}")
        try:
            df = self.identifier()
        except Exception as e:
            logger.error(traceback.format_exc())
            # message = str(e) + ',风机编号:' + self.wind_turbine_number
            # raise Exception('打标签失败:' + message)
            self.df.loc[:, 'lab'] = -999
            return self.df
        logger.info(f"打标签结束,{df.shape},耗时:{datetime.datetime.now() - begin}")
        return df