You can not select more than 25 topics
Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.
286 lines
11 KiB
286 lines
11 KiB
2 weeks ago
|
# -*- coding: utf-8 -*-
|
||
|
|
"""
|
||
|
|
Created on Sun Feb 28 10:04:26 2016
|
||
|
|
PCA source code————最新更新————————————————————————————
|
||
|
|
@author:
|
||
|
|
"""
|
||
|
|
|
||
|
|
import time
|
||
|
|
import numpy as np
|
||
|
|
import pandas as pd
|
||
|
|
from scipy.stats import norm
|
||
|
|
from scipy.stats.distributions import chi2
|
||
|
|
import json
|
||
|
|
import sys
|
||
|
|
import pymssql
|
||
|
|
import requests
|
||
|
|
import datetime
|
||
|
|
from scipy.stats import norm
|
||
|
|
from scipy.stats import f
|
||
|
|
from scipy.stats import chi2
|
||
|
|
import jenkspy
|
||
|
|
import xlrd
|
||
|
|
from recon import Lars, recon_fault_diagnosis_r, recon_fault_diagnosis_r_l, recon_fault_diagnosis_r_c
|
||
|
|
|
||
|
|
|
||
|
|
"""
|
||
|
|
参数：
|
||
|
|
- XMat：传入的是一个numpy的矩阵格式，行表示样本数，列表示特征
|
||
|
|
- k：表示取前k个特征值对应的特征向量
|
||
|
|
返回值：
|
||
|
|
- finalData：参数一指的是返回的低维矩阵，对应于输入参数二
|
||
|
|
- reconData：参数二对应的是移动坐标轴后的矩阵
|
||
|
|
"""
|
||
|
|
|
||
|
|
|
||
|
|
|
||
|
|
|
||
|
|
|
||
|
|
class MSSQL:
|
||
|
|
def __init__(self, host, user, pwd, database):
|
||
|
|
self.host = host
|
||
|
|
self.user = user
|
||
|
|
self.pwd = pwd
|
||
|
|
self.db = database
|
||
|
|
|
||
|
|
def __GetConnect(self):
|
||
|
|
"""
|
||
|
|
得到连接信息
|
||
|
|
返回: conn.cursor()
|
||
|
|
"""
|
||
|
|
if not self.db:
|
||
|
|
raise (NameError, "没有设置数据库信息")
|
||
|
|
self.conn = pymssql.connect(host=self.host, user=self.user, password=self.pwd, database=self.db, charset="utf8")
|
||
|
|
cur = self.conn.cursor()
|
||
|
|
if not cur:
|
||
|
|
raise (NameError, "连接数据库失败")
|
||
|
|
else:
|
||
|
|
return cur
|
||
|
|
|
||
|
|
def ExecQuery(self, sql):
|
||
|
|
"""
|
||
|
|
执行查询语句
|
||
|
|
返回的是一个包含tuple的list，list的元素是记录行，tuple的元素是每行记录的字段
|
||
|
|
"""
|
||
|
|
cur = self.__GetConnect()
|
||
|
|
cur.execute(sql)
|
||
|
|
resList = cur.fetchall()
|
||
|
|
|
||
|
|
# 查询完毕后必须关闭连接
|
||
|
|
self.conn.close()
|
||
|
|
return resList
|
||
|
|
|
||
|
|
def ExecNonQuery(self, sql):
|
||
|
|
"""
|
||
|
|
执行非查询语句
|
||
|
|
|
||
|
|
调用示例：
|
||
|
|
cur = self.__GetConnect()
|
||
|
|
cur.execute(sql)
|
||
|
|
self.conn.commit()
|
||
|
|
self.conn.close()
|
||
|
|
"""
|
||
|
|
cur = self.__GetConnect()
|
||
|
|
cur.execute(sql)
|
||
|
|
self.conn.commit()
|
||
|
|
self.conn.close()
|
||
|
|
|
||
|
|
|
||
|
|
def get_model_by_ID(model_id):
|
||
|
|
ms = MSSQL(host="192.168.10.250", user="sa", pwd="powerSIS#123", database="alert")
|
||
|
|
resList = ms.ExecQuery("SELECT Model_info FROM [alert].[dbo].[Model_CFG] where \"model_id\"=" + str(model_id))
|
||
|
|
# return json.loads(resList[0][0])["para"]
|
||
|
|
return json.loads(resList[0][0])
|
||
|
|
|
||
|
|
|
||
|
|
def pca(model, LockVariable, Data_origin):
|
||
|
|
Data = (Data_origin - model["Train_X_mean"]) / model["Train_X_std"]
|
||
|
|
featValue = np.array(model["featValue"]) # 训练数据的特征值
|
||
|
|
featVec = np.array(model["featVec"]) # 训练数据的特征向量
|
||
|
|
k = (model["K"]) # 主元个数
|
||
|
|
# selectVec = np.array(model["selectVec"])
|
||
|
|
selectVec = featVec[:, 0:k] ####自己选择的
|
||
|
|
# index = np.argsort(-np.array(featValue)) # 按照featValue进行从大到小排序
|
||
|
|
featValue_sort = featValue # [index] # 排序后的特征值
|
||
|
|
'''
|
||
|
|
featValue, featVec = np.linalg.eig(model["COV"]) # 求解协方差矩阵的特征值和特征向量
|
||
|
|
|
||
|
|
index = np.argsort(-featValue) # 按照featValue进行从大到小排序
|
||
|
|
featValue_sort =featValue[index]#排序后的特征值
|
||
|
|
selectVec = np.matrix(featVec[:,index[:int(model["K"])]]) # 所以这里需要进行转置P
|
||
|
|
'''
|
||
|
|
############----------*********-SPE-**************----------########################
|
||
|
|
numbel_variable = featValue.shape[0]
|
||
|
|
# LockVariable="3,5"
|
||
|
|
|
||
|
|
C_ = np.eye(numbel_variable) - np.dot(selectVec, selectVec.T)
|
||
|
|
X_SPE = C_.T
|
||
|
|
D_SPE = C_
|
||
|
|
DIAG_SPE = np.eye(numbel_variable)
|
||
|
|
|
||
|
|
'''
|
||
|
|
# ************************调用LARS*******************************
|
||
|
|
t = 50000
|
||
|
|
lamuta = 1
|
||
|
|
limit_line = model["QCUL_99"]
|
||
|
|
beta_path=[]
|
||
|
|
for i in range(Data.shape[0]):
|
||
|
|
Y=Data[i,:]
|
||
|
|
beta, mse=Lars(X_SPE, Y, D_SPE, DIAG_SPE, t, limit_line, lamuta)
|
||
|
|
beta_end=abs(beta[-1,:])
|
||
|
|
jenk=jenkspy.jenks_breaks(beta_end,5)
|
||
|
|
limt=(jenk[1]+jenk[2])/2
|
||
|
|
index=np.where(beta_end>limt)[0]
|
||
|
|
beta_path.append(beta[-1,:])
|
||
|
|
'''
|
||
|
|
############----------*********-T2-**************----------########################
|
||
|
|
DIAG_T2 = np.linalg.pinv(np.diag(featValue_sort[:int(model["K"])]))
|
||
|
|
D_T2 = selectVec.copy()
|
||
|
|
X_T2 = np.dot(D_T2, np.linalg.cholesky(DIAG_T2)).T
|
||
|
|
|
||
|
|
############----------*********-综合指标-**************----------########################
|
||
|
|
II = featValue_sort.copy()
|
||
|
|
II[:int(model["K"])] = II[:int(model["K"])] * model["T2CUL_99"]
|
||
|
|
II[int(model["K"]):] = model["QCUL_99"]
|
||
|
|
DIAG_Fai = np.linalg.inv(np.diag(II))
|
||
|
|
D_Fai = featVec.copy()
|
||
|
|
X_Fai = np.dot(D_Fai, np.linalg.cholesky(DIAG_Fai)).T
|
||
|
|
# ************************调用LARS*******************************
|
||
|
|
t = 50000
|
||
|
|
lamuta = 1
|
||
|
|
# limit_line = model["Kesi_99"]/np.sqrt(numbel_variable)#修改
|
||
|
|
limit_line = model["Kesi_99"]*1
|
||
|
|
beta_path = []
|
||
|
|
SPE_list = []
|
||
|
|
FAI_list = []
|
||
|
|
paraState = np.zeros([np.array(Data_origin).shape[0], np.array(Data_origin).shape[1]])
|
||
|
|
if Data.shape[1] >= 12:
|
||
|
|
para_length = 3
|
||
|
|
elif 12 > Data.shape[1] >= 7:
|
||
|
|
para_length = 2
|
||
|
|
else:
|
||
|
|
para_length = 1
|
||
|
|
plots_matrix = [] # 贡献图法的矩阵
|
||
|
|
plots_index = [] # 贡献图法的index
|
||
|
|
# para_length = Data.shape[1]
|
||
|
|
for i in range(Data.shape[0]):
|
||
|
|
Y = Data[i, :] # 测试数据的每一行
|
||
|
|
if i == 1051:
|
||
|
|
print("aaa")
|
||
|
|
#########*********************计算SPE******************************
|
||
|
|
SPE_line = np.dot(Y, C_).dot(Y.T)
|
||
|
|
SPE_list.append(SPE_line)
|
||
|
|
#########################计算综合指标##########
|
||
|
|
temp = np.dot(Y.T, D_Fai).dot(DIAG_Fai).dot(D_Fai.T).dot(Y)
|
||
|
|
if temp > 100:
|
||
|
|
print("aa")
|
||
|
|
FAI_list.append(np.dot(Y.T, D_Fai).dot(DIAG_Fai).dot(D_Fai.T).dot(Y))
|
||
|
|
# **************计算LARS***************
|
||
|
|
beta, mse = Lars(X_Fai, Y, D_Fai, DIAG_Fai, t, limit_line, LockVariable)
|
||
|
|
beta_end = abs(beta[-1, :])
|
||
|
|
pi = len(beta_end)
|
||
|
|
if pi > 7:
|
||
|
|
jenk = jenkspy.jenks_breaks(beta_end, 5)
|
||
|
|
else:
|
||
|
|
jenk = jenkspy.jenks_breaks(beta_end, 2)
|
||
|
|
limt = (jenk[1] + jenk[2]) / 2
|
||
|
|
index = np.where(beta_end > 0)[0]
|
||
|
|
if len(index) > para_length:
|
||
|
|
# 如果Lars诊断出来的故障个数大于限值则进一步重构
|
||
|
|
res = recon_fault_diagnosis_r_c(Y, D_Fai @ DIAG_Fai @ D_Fai.T, limit_line, list(zip(index, beta_end[index])), model,
|
||
|
|
True, X_SPE @ X_SPE.T, LockVariable, selectVec, rbc=None)
|
||
|
|
if not isinstance(res[0], list):
|
||
|
|
# 如果返回的res[0]是array类型则说明重构有结果,如果没有重构出来则返回的是[]
|
||
|
|
# 如果是采用的贡献图法进行重构则res[1]为plot,否则是array类型,表示故障参数的index
|
||
|
|
if res[1] == "plot":
|
||
|
|
# beta[-1, :] = res[0]
|
||
|
|
plots_matrix.append(res[0])
|
||
|
|
plots_index.append(i)
|
||
|
|
else:
|
||
|
|
beta[-1, :], index = res[0].T, res[1]
|
||
|
|
elif len(index) <= para_length and len(index) != 0:
|
||
|
|
res = recon_fault_diagnosis_r_l(Y, D_Fai @ DIAG_Fai @ D_Fai.T, index)
|
||
|
|
beta[-1, :], index = res[0].T, res[1]
|
||
|
|
paraState[i, index] = 1
|
||
|
|
beta_new = beta[-1, :] * paraState[i, :]
|
||
|
|
beta_path.append(beta_new)
|
||
|
|
beta_path = np.array(beta_path)
|
||
|
|
# ###############-------------------------------------------------------------###############
|
||
|
|
# finalData = np.dot(Data - beta_path, selectVec).dot(selectVec.T)
|
||
|
|
finalData = Data - beta_path
|
||
|
|
reconData = np.add(np.multiply(finalData, model["Train_X_std"]), model["Train_X_mean"]) # 重构值
|
||
|
|
if plots_index:
|
||
|
|
# 如果不为[],则说明重构的时候采用了贡献图法,则需要将对应的重构值替换成贡献图法的重构结果.
|
||
|
|
reconData[plots_index] = plots_matrix
|
||
|
|
errorData = Data_origin - reconData # 偏差值
|
||
|
|
|
||
|
|
# cos检验值
|
||
|
|
R = 0
|
||
|
|
for index in range(0, reconData.shape[1]):
|
||
|
|
vector1 = Data_origin[:, index]
|
||
|
|
vector2 = np.array(reconData)[:, index]
|
||
|
|
R += np.dot(vector1, vector2.T) / (np.sqrt(np.sum(vector1 ** 2)) * np.sqrt(np.sum(vector2 ** 2)))
|
||
|
|
R /= reconData.shape[1]
|
||
|
|
items = [('sampleData',np.transpose(np.array(Data_origin)).tolist()), ('reconData', reconData.T.tolist())
|
||
|
|
, ('errorData', errorData.T.tolist()), ('R', R.tolist()), ('SPE', FAI_list),
|
||
|
|
('paraState', paraState.tolist())]
|
||
|
|
result = dict(items) # json.dumps(result)
|
||
|
|
return result
|
||
|
|
|
||
|
|
|
||
|
|
def get_history_value(points, time, interval):
|
||
|
|
# url="http://192.168.1.201:8080/openPlant/getMultiplePointHistorys"
|
||
|
|
url = "http://10.35.2.238:9000/exawebapi/exatime/GetSamplingValueArrayFloat"
|
||
|
|
headers = {"Content-Type": "application/json;charset=utf-8"} # ,"token":get_token()
|
||
|
|
point_array = points.split(",")
|
||
|
|
time_span = time.split(";")
|
||
|
|
value_array = []
|
||
|
|
for item in point_array:
|
||
|
|
for time_piece in time_span:
|
||
|
|
st = time_piece.split(",")[0]
|
||
|
|
et = time_piece.split(",")[1]
|
||
|
|
para = {"ItemName": item, "StartingTime": st, "TerminalTime": et, "SamplingPeriod": interval}
|
||
|
|
response = requests.get(url, headers=headers, params=para)
|
||
|
|
content = response.text.replace('"[', '[').replace(']"', ']')
|
||
|
|
value = json.loads(content)
|
||
|
|
value_group = []
|
||
|
|
for row in value:
|
||
|
|
value_group.append(row[1])
|
||
|
|
value_array.append(value_group)
|
||
|
|
return np.transpose(np.array(value_array))
|
||
|
|
|
||
|
|
|
||
|
|
if __name__ == "__main__":
|
||
|
|
|
||
|
|
info_str = '{"Test_Data":{"time":"2021-07-18 10:34:16,2021-07-19 10:34:16","points":"JL_D2_20DAS05A:LBS60CP101.PNT,JL_D2_20DAS06A:LCJ60CT301.PNT,JL_D2_20DAS06A:LCA30CT305.PNT,JL_D2_20DAS06A:LCA30CT303.PNT,JL_D2_20DAS06A:LCA30CT304.PNT,JL_D2_20DAS06A:LCJ51CT301.PNT","interval":300000,"AddBias":[0.0,0.0,0.0,0.0,0.0,0.0],"AddBias_Time":"2021-07-18 10:34:16,2021-07-19 10:34:16"},"Model_id":770,"version":"v-2021-04-02 10:26:32"}'
|
||
|
|
info = json.loads(info_str)
|
||
|
|
model_id = info["Model_id"]
|
||
|
|
version = info["version"]
|
||
|
|
import PCA_Test
|
||
|
|
if version == "v-test":
|
||
|
|
res = PCA_Test.get_model_by_ID(model_id, version)
|
||
|
|
else:
|
||
|
|
res = PCA_Test.get_model_by_id_and_version(model_id, version)
|
||
|
|
filename = res["algorithm"]
|
||
|
|
if filename == "PCA":
|
||
|
|
lock = []
|
||
|
|
point_info = res["pointInfo"]
|
||
|
|
for i in range(len(point_info)):
|
||
|
|
try:
|
||
|
|
if point_info[i]["lock"]:
|
||
|
|
lock.append(i)
|
||
|
|
except:
|
||
|
|
continue
|
||
|
|
lock = []
|
||
|
|
Test_Data = info["Test_Data"]
|
||
|
|
points = Test_Data["points"]
|
||
|
|
time1 = Test_Data["time"]
|
||
|
|
interval = Test_Data["interval"]
|
||
|
|
model = res["para"]["Model_info"]
|
||
|
|
Data = get_history_value(points, time1, interval)
|
||
|
|
t1 = time.time()
|
||
|
|
result = pca(model, lock, Data)
|
||
|
|
t2 = time.time()
|
||
|
|
print(t2 - t1)
|
||
|
|
# print(result)
|