WTOAAM/dataAnalysisBusiness/algorithm/temperatureEnvironmentAnalyst.py

import os
import pandas as pd
import numpy as np
import pandas as pd
import matplotlib.pyplot as plt
import seaborn as sns
import plotly.graph_objects as go
from plotly.subplots import make_subplots
from geopy.distance import geodesic
from behavior.analyst import Analyst
from utils.directoryUtil import DirectoryUtil as dir
from algorithmContract.confBusiness import *
import common.turbineInfo as turbineInfo


class TemperatureEnvironmentAnalyst(Analyst):
    """
    风电机组大部件温升分析
    """

    def typeAnalyst(self):
        return "temperature_environment"

    def turbinesAnalysis(self, dataFrameMerge,outputAnalysisDir, confData: ConfBusiness):        
        # 检查所需列是否存在
        required_columns = {confData.field_env_temp,
                            Field_NameOfTurbine}
        if not required_columns.issubset(dataFrameMerge.columns):
            raise ValueError(f"DataFrame缺少必要的列。需要的列有: {required_columns}")
        
        turbineInfos = turbineInfo.loadTurbineInfo(confData.turbineInfoFilePathCSV)

        #  环境温度 分析
        turbineEnvTempData =dataFrameMerge.groupby(Field_NameOfTurbine).agg(
            {confData.field_env_temp : 'median'}).reset_index(names=[Field_NameOfTurbine]) 
        mergeData = self.merge_Data(Field_NameOfTurbine,turbineInfos, turbineEnvTempData,confData)
        
        self.draw(mergeData,outputAnalysisDir, confData)
    

    def merge_Data(self,fieldTurbineName, turbineInfos:pd.DataFrame, turbineEnvTempData, confData: ConfBusiness):
        """
        将每台机组的环境温度均值数据与机组信息，按机组合并

        参数:
        turbineInfos (pandas.DataFrame): 机组信息数据
        turbineEnvTempData (pandas.DataFrame): 每台机组的环境温度均值数据

        返回:
        pandas.DataFrame: 每台机组的环境温度均值数据与机组信息合并数据
        """

        """
        合并类型how的选项包括：

        'inner': 内连接，只保留两个DataFrame中都有的键的行。
        'outer': 外连接，保留两个DataFrame中任一或两者都有的键的行。
        'left': 左连接，保留左边DataFrame的所有键，以及右边DataFrame中匹配的键的行。
        'right': 右连接，保留右边DataFrame的所有键，以及左边DataFrame中匹配的键的行。
        """
        # turbineEnvTempData[fieldTurbineName]=turbineEnvTempData[fieldTurbineName].astype(str)
        merge_data = pd.merge(turbineInfos, turbineEnvTempData, on=[fieldTurbineName], how='inner')

        return merge_data

    # 定义查找给定半径内点的函数
    def find_points_within_radius(self, data, center, field_temperature_env, radius):
        points_within_radius = []
        for index, row in data.iterrows():
            distance = geodesic(
                (center[2], center[1]), (row['latitude'], row['longitude'])).meters
            if distance <= radius:
                points_within_radius.append(
                    (row['turbine_name'], row[field_temperature_env]))
        return points_within_radius

    fieldTemperatureDiff="temperature_diff"
    def draw(self,dataFrame : pd.DataFrame, outputAnalysisDir,confData: ConfBusiness,charset=charset_unify):
        # 处理数据
        dataFrame['new'] = dataFrame.loc[:, [Field_NameOfTurbine , 'longitude', 'latitude', confData.field_env_temp]].apply(tuple, axis=1)
        coordinates = dataFrame['new'].tolist()
        df = pd.DataFrame(coordinates, columns=[Field_NameOfTurbine, 'longitude', 'latitude', confData.field_env_temp])

        # 查找半径内的点
        points_within_radius = {coord: self.find_points_within_radius(dataFrame,coord,confData.field_env_temp,confData.rotor_diameter*10) for coord in coordinates}
        res = []
        for center, nearby_points in points_within_radius.items():
            current_temp = dataFrame[dataFrame[Field_NameOfTurbine] == center[0]][confData.field_env_temp].iloc[0]
            target_tuple = (center[0], current_temp)
            if target_tuple in nearby_points:
                nearby_points.remove(target_tuple)
            mean_temp = np.median([i[1] for i in nearby_points]) if nearby_points else current_temp
            res.append((center[0], nearby_points, mean_temp, current_temp))
        res = pd.DataFrame(res, columns=[Field_NameOfTurbine, '周边机组', '周边机组温度', '当前机组温度'])
        res[self.fieldTemperatureDiff] = res['当前机组温度'] - res['周边机组温度']

        fig, ax = plt.subplots(figsize=(16,8),dpi=96)
        # 设置x轴刻度值旋转角度为45度  
        plt.tick_params(axis='x', rotation=45)

        sns.barplot(x=Field_NameOfTurbine,y=self.fieldTemperatureDiff,data=res,ax=ax,color='dodgerblue')
        plt.axhline(y=5,ls=":",c="red")#添加水平直线
        plt.axhline(y=-5,ls=":",c="red")#添加水平直线
        ax.set_ylabel('temperature_difference')
        ax.set_title('temperature Bias')
        plt.savefig(outputAnalysisDir +'//'+ "{}环境温差Bias.png".format(confData.farm_name),bbox_inches='tight',dpi=120)


        fig2, ax2 = plt.subplots(figsize=(16,8),dpi=96)
        # 设置x轴刻度值旋转角度为45度  
        plt.tick_params(axis='x', rotation=45)
        
        sns.barplot(x=Field_NameOfTurbine ,y='当前机组温度',data=res,ax=ax2,color='dodgerblue')
        ax2.set_ylabel('temperature')
        ax2.set_title('temperature median')
        plt.savefig(outputAnalysisDir +'//'+ "{}环境温度均值.png".format(confData.farm_name),bbox_inches='tight',dpi=120)