CN115358463B - Ecological sensitive area power transmission and transformation construction engineering water environment monitoring and influence assessment method - Google Patents

Ecological sensitive area power transmission and transformation construction engineering water environment monitoring and influence assessment method Download PDF

Info

Publication number
CN115358463B
CN115358463B CN202210996244.4A CN202210996244A CN115358463B CN 115358463 B CN115358463 B CN 115358463B CN 202210996244 A CN202210996244 A CN 202210996244A CN 115358463 B CN115358463 B CN 115358463B
Authority
CN
China
Prior art keywords
water environment
power transmission
input
index
output
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active
Application number
CN202210996244.4A
Other languages
Chinese (zh)
Other versions
CN115358463A (en
Inventor
杨波
杨东俊
陈曦
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Changsha University
Original Assignee
Changsha University
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Changsha University filed Critical Changsha University
Priority to CN202210996244.4A priority Critical patent/CN115358463B/en
Publication of CN115358463A publication Critical patent/CN115358463A/en
Application granted granted Critical
Publication of CN115358463B publication Critical patent/CN115358463B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N33/00—Investigating or analysing materials by specific methods not covered by groups G01N1/00 - G01N31/00
    • G01N33/18—Water
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N33/00—Investigating or analysing materials by specific methods not covered by groups G01N1/00 - G01N31/00
    • G01N33/18—Water
    • G01N33/1806—Biological oxygen demand [BOD] or chemical oxygen demand [COD]
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/04—Architecture, e.g. interconnection topology
    • G06N3/049—Temporal neural networks, e.g. delay elements, oscillating neurons or pulsed inputs
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/08—Learning methods
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
    • G06Q50/06—Energy or water supply
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
    • G06Q50/10—Services
    • G06Q50/26—Government or public services
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y04—INFORMATION OR COMMUNICATION TECHNOLOGIES HAVING AN IMPACT ON OTHER TECHNOLOGY AREAS
    • Y04S—SYSTEMS INTEGRATING TECHNOLOGIES RELATED TO POWER NETWORK OPERATION, COMMUNICATION OR INFORMATION TECHNOLOGIES FOR IMPROVING THE ELECTRICAL POWER GENERATION, TRANSMISSION, DISTRIBUTION, MANAGEMENT OR USAGE, i.e. SMART GRIDS
    • Y04S10/00—Systems supporting electrical power generation, transmission or distribution
    • Y04S10/50—Systems or methods supporting the power network operation or management, involving a certain degree of interaction with the load-side end user applications

Landscapes

  • Engineering & Computer Science (AREA)
  • Business, Economics & Management (AREA)
  • Health & Medical Sciences (AREA)
  • Physics & Mathematics (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Theoretical Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • General Health & Medical Sciences (AREA)
  • Economics (AREA)
  • Tourism & Hospitality (AREA)
  • Chemical & Material Sciences (AREA)
  • Strategic Management (AREA)
  • Human Resources & Organizations (AREA)
  • Biomedical Technology (AREA)
  • Marketing (AREA)
  • General Business, Economics & Management (AREA)
  • Molecular Biology (AREA)
  • Computational Linguistics (AREA)
  • Medicinal Chemistry (AREA)
  • Pathology (AREA)
  • Primary Health Care (AREA)
  • Immunology (AREA)
  • Development Economics (AREA)
  • Artificial Intelligence (AREA)
  • Biochemistry (AREA)
  • Biophysics (AREA)
  • Analytical Chemistry (AREA)
  • Data Mining & Analysis (AREA)
  • Evolutionary Computation (AREA)
  • Food Science & Technology (AREA)
  • Computing Systems (AREA)
  • General Engineering & Computer Science (AREA)
  • Mathematical Physics (AREA)
  • Software Systems (AREA)
  • Operations Research (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Water Supply & Treatment (AREA)
  • Quality & Reliability (AREA)
  • Public Health (AREA)
  • Educational Administration (AREA)

Abstract

The invention relates to the field of power transmission and transformation environment influence evaluation, in particular to a method for monitoring and evaluating the water environment of power transmission and transformation construction engineering in an ecology sensitive area; according to the method, a time sequence of the index change along with time is constructed by monitoring the water environment index of the sewage receiving body, meanwhile, hydrologic indexes such as flow rate, water temperature and water level of the sewage receiving body at corresponding moments are sampled to form a multi-index vector historical time sequence, a nonlinear mapping network is constructed by learning historical sample data of the time sequence, the multi-index vector value at the next moment is predicted according to the time sequence, and whether the environmental influence and the evaluation influence degree of the sewage receiving body are caused by the power transmission and transformation construction engineering or not is judged by checking the deviation degree of the multi-index vector predicted value and the water environment monitoring threshold value of the ecological sensitive area, so that an effective technical means is provided for accurately monitoring and protecting the water environment of the power transmission and transformation construction engineering of the ecological sensitive area.

Description

Ecological sensitive area power transmission and transformation construction engineering water environment monitoring and influence assessment method
Technical Field
The invention relates to the field of power transmission and transformation environment influence evaluation, in particular to a method for monitoring and evaluating the water environment and influence of power transmission and transformation construction engineering in an ecology sensitive area.
Background
The power transmission and transformation project is a collective term for power transmission line construction and transformer installation projects. The evaluation of the power transmission and transformation environment influence is a necessary requirement for realizing national strategy and ecological environment protection planning, is an important means for cooperatively promoting economic high-quality development and ecological environment high-level protection, and is a first defense line in development. The power transmission and transformation project can be divided into an ac power transmission and transformation project and a dc power transmission project, wherein the ac power transmission and transformation project comprises a power transmission line and a transformer substation (or a switching station and a series compensation station), and the dc power transmission project comprises a power transmission line, a converter station and a grounding electrode system. According to the regulations of the national environmental protection standard ' environmental impact evaluation technology guide ' of China, when a power transmission and transformation construction project enters the ring regulated by the ' construction project environmental impact evaluation classification management directoryWhen the environment is sensitive, ecological planning compliance, environment rationality and feasibility analysis of construction projects should be carried out. The influence of power transmission and transformation engineering on water environment mainly comprises two aspects: firstly, construction wastewater, including foundation construction slurry, construction site mixing wastewater, sand and stone material flushing and material cleaning and screening wastewater, machine repairing oil-containing wastewater and the like; and secondly, domestic sewage. The influence of power transmission and transformation engineering on the surface water environment is mainly represented by the following indexes: the first is the pH value, which is the negative value of the usual logarithm of the concentration of hydrogen ions in aqueous solutions, i.e. -lg [ H+]A representation; secondly, the chemical oxygen demand COD (Chemical Oxygen Demand), namely, the amount of the reducing substances to be oxidized in the water sample is measured by a chemical method, and the pollution degree of the reducing substances in the water is reflected; thirdly, the biochemical oxygen demand BOD (Biochemical Oxygen Demand), namely the amount of dissolved oxygen consumed in the biochemical reaction process of decomposing the biochemically degradable organic matters in the water by the microorganism under certain conditions, reflects the content of the organic matters and other aerobic pollutants in the water, and is called as five-day Biochemical Oxygen Demand (BOD) if the biological oxidation is carried out for five days 5 ) The method comprises the steps of carrying out a first treatment on the surface of the Fourthly, ammonia nitrogen NH 3 N, an indicator of ammonia nitrogen content in water, is a direct factor responsible for water eutrophication; fifthly, petroleum is mainly from pollution of wastewater and domestic sewage.
The ecological sensitive area is used as an area which has special sensitivity to human production and living activities and is extremely easy to be influenced by artificial improper development activities to generate ecological negative effects, and has higher requirements on the water environment influence of power transmission and transformation construction engineering. According to the environmental protection standard HJ 24-2020 of China, the water environment influence evaluation factors of the power transmission and transformation construction project comprise pH value, COD and BOD 5 、NH 3 -N and petroleum. In order to control the influence of power transmission and transformation construction engineering on the water environment of an ecologically sensitive area, the installation of sewage treatment equipment, sewage recycling facilities, a rain and sewage diversion pipe network, side station supervision sewage treatment facilities, side station supervision transformer station accident oil ponds, concrete stirring wastewater treatment facilities and the like is required. However, due to the reasons of irregular management of owners of power transmission and transformation construction engineering, insufficient work of environmental supervision units, improper handling of environmental accidents and the like, the power transmission and transformation construction engineering is especially suitable for crossing water bodies andin engineering construction of wiring near the water body, the water environment of the sewage receiving body can be seriously influenced, and irreversible destructive effect is generated on the water environment of the ecologically sensitive area. Therefore, the method has important significance in monitoring and evaluating the influence of the water environment of the power transmission and transformation construction engineering in the ecologically sensitive area.
Disclosure of Invention
The invention aims to provide a water environment monitoring and influence assessment method for power transmission and transformation construction engineering in an ecological sensitive area, which monitors the water environment index of a sewage receiving body (the pH value, COD and BOD which are monitored by regulations of power transmission and transformation engineering of environmental protection standard of national environmental protection technology guide rule of China) 5 、NH 3 -N, petroleum and the like), constructing a time sequence of the index change along with time, sampling hydrologic indexes such as the flow rate, the water temperature, the water level and the like of the sewage receiving body at corresponding moments, forming a multi-index vector (a water-containing environment index and a hydrologic index) historical time sequence, constructing a nonlinear mapping network through learning historical sample data of the time sequence, predicting multi-index vector values at the next moment according to the nonlinear mapping network, and judging whether the power transmission and transformation construction engineering has environmental influence and evaluation influence degree on the sewage receiving body or not by checking the deviation degree of a multi-index vector predicted value and an ecology sensitive area water environment monitoring threshold value, thereby providing an effective technical means for the precise monitoring and protection of the power transmission and transformation construction engineering water environment in the ecology sensitive area.
In order to achieve the above purpose, the technical solution adopted by the invention is as follows: the method for monitoring and evaluating the water environment and the influence of the power transmission and transformation construction project in the ecological sensitive area comprises the following steps:
s1, acquiring water environment indexes of a power transmission and transformation construction engineering sewage receiving body at a monitoring section and a sampling point, wherein the water environment indexes comprise pH value, COD and BOD 5 、NH 3 -N and petroleum, etc., constituting a water environment index time series: x is X t ＝ {x 1，t ，x 2，t ，x 3，t ，x 4，t ，x 5，t }＝{x l，t |l＝[1，2，…，5]T is the sampling time, t.epsilon.1, 2, …, T]T represents the maximum sampling time, l is the first water environment index x l，t Is a woven fabric of (a)Number, x 1，t The pH value is represented, and the method is dimensionless; x is x 2，t Represents chemical oxygen demand, COD, in milligrams per liter (mg/L); x is x 3，t Representing the BOD of five days 5 Milligrams per liter (mg/L); x is x 4，t Representing ammonia nitrogen content NH 3 -N in milligrams per liter (mg/L); x is x 5，t Indicating petroleum in milligrams per liter (mg/L).
S2, acquiring hydrologic indexes of the power transmission and transformation construction engineering sewage receiving body at the monitoring section and the sampling point, wherein the hydrologic indexes comprise water level, water temperature, flow speed and the like, and a hydrologic index time sequence is formed: y is Y t ＝{H t ，C t ，V t T is the sampling time, t.epsilon.1, 2, …, T]，H t Representing the water level in meters (m); c (C) t Temperature in degrees celsius (°c); v (V) t The water flow rate is expressed in meters per second (m/s).
S3, constructing a power transmission and transformation construction engineering sewage body multi-index vector (water-containing environment index and hydrologic index) historical time sequence through the water environment index time sequence and the hydrologic index time sequence:
M＝{M t |t∈[1，2，…，T]}
wherein M is t ＝{X′ t ，Y t }，X′ t Is X t Is a subset of (i.e.)
Figure SMS_1
X′ t The pH value, COD and BOD of the factors are evaluated by the influence of water environment 5 、NH 3 And (3) one or more indexes of N, petroleum and the like, wherein the number of the specific indexes is comprehensively determined according to the technical conditions of acquisition and monitoring and the environmental protection supervision requirements of the ecological sensitive area.
S4, according to the multi-index vector historical time sequence M= { M t |t∈[1，2，…，T]Building an input-output map. Let N be the input dimension of the input-output map and N < T, then the multi-index vector historical time series M can construct T-N input-output maps:
(M 1 ，M 2 ，…，M N )→M N+1
(M 2 ，M 3 ，…，M N+1 )→M N+2
(M T-N ，M T-N+1 ，…，M T-1 )→M T
thus, the above-described input-output mapping is as follows:
(M s ，M s+1 ，…，M s+N-1 )→M s+N ，s∈[1，2，…，T-N]
s represents the number of the s-th input-output mapping, and the value is [1,2, …, T-N]An integer therebetween. (M) s ，M s+1 ，…，M s+N-1 ) Representing the input of the s-th input-output mapping, M s+N Representing the expected output of the s-th input-output map.
S5, constructing a power transmission and transformation construction engineering water environment index nonlinear mapping network prediction model based on a multi-index vector historical time sequence, wherein the input of the model is the input of an input-output mapping, and the output is a multi-index vector predicted value at a predicted moment.
The nonlinear mapping network prediction model adopts a layered structure of an input layer, a hidden layer and an output layer. In the input layer, the input of the nonlinear mapping network prediction model is the input of the s-th input-output mapping (M s ，M s+1 ，…，M s+N-1 ) The method comprises the steps of carrying out a first treatment on the surface of the In the output layer, the output of the model is the multi-index vector predicted value of the (s+N) th predicted time
Figure SMS_2
In the hidden layer, the node number of the hidden layer is equal to the input dimension N of the input-output mapping, and the nth node is marked as U n ，n∈[1，2，…，N]. For the nth node, four neurons f, i, g and o are adopted to realize the coordinated memory of long-term trend and short-term fluctuation of the time sequence, and the corresponding neuron outputs are as follows:
f n ＝σ(w f Z n +R f H n-1 +b f )
i n ＝σ(w i Z n +R i H n-1 +b i )
g n ＝δ(w g Z n +R g H n-1 +b g )
o n ＝σ(w o Z n +R o H n-1 +b o )
wherein f n ，i n ，g n And o n The excitation outputs of the four neurons f, i, g and o, σ (&) and δ (&) are sigmoid and tanh functions, b, respectively f ，b i ，b g And b o Deviations of four neurons, f, i, g and o, respectively, w f ，w i ， w g And w o Input weights of four neurons of f, i, g and o, respectively, R f ，R i ，R g And R is o Regression weights for the four neurons f, i, g and o, respectively. According to the model, the nth node U n Long-term memory C n And short-term memory H n The calculation is as follows:
Figure SMS_3
Figure SMS_4
wherein H is n-1 And C n-1 Characterizing the n-1 th node U respectively n-1 Short-term memory and long-term memory of (a). When n=n, then there are:
Figure SMS_5
this means the nth node U N Output H of (2) N Can be obtained by iterative calculation. H at this time N I.e. the multi-index vector predicted value of the s+N predicted time
Figure SMS_6
And S6, training a nonlinear mapping network prediction model of the water environment index of the power transmission and transformation construction engineering according to the T-N input-output mappings, and identifying parameters of the model. The method comprises the following steps:
in S4, according to the multi-index vector history time sequence M= { M t |t∈[1，2，…，T]T-N input-output maps are constructed:
(M s ，M s+1 ，…，M s+N-1 )→M s+N ，s∈[1，2，…，T-N]
s5, according to the power transmission and transformation construction engineering water environment index nonlinear mapping network prediction model, setting the input of the model as the input of the input-output mapping (M s ，M s+1 ，…，M s+N-1 ) The output of the model is the multi-index vector predicted value of the (s+N) th predicted time
Figure SMS_7
The parameters of the model to be identified are W:
W＝{W i ，W r ，b}
W i ＝{w f ，w i ，w g ，w o }
W r ＝{R f ，R i ，R g ，R o }
b＝{b f ，b i ，b g ，b o }
wherein W is i Input weights w representing four neurons f, i, g and o f ，w i ，w g And w o Is a collection of (3); w (W) r Regression weights R representing four neurons f, i, g and o f ，R i ，R g And R is o B represents the bias b of four neurons f, i, g and o f ，b i ，b g And b o Is a set of (3).
The identification process of the model parameter W is as follows:
first calculate the training error E of the model train ：
Figure SMS_8
The loss function J (W) of the model is determined as:
Figure SMS_9
alpha is an adjustment factor for optimizing training error E in preference during iterative calculation train The trade-off is made between (first term to the right of the equal sign of the J (W) expression) and the priority optimization model parameter W (second term to the right of the equal sign of the J (W) expression). Alpha typically takes a value of 0.1.
The model parameters W are then determined using iterative calculations:
Figure SMS_10
beta is the learning rate in the iterative calculation, ensuring that the variation of W in each iterative calculation is within a suitable range. Beta is generally taken to be 0.001.
S7, predicting the multi-index vector predicted value at the time T+1 by adopting a power transmission and transformation construction engineering water environment index nonlinear mapping network prediction model according to the finally determined model parameter W in the S6. Setting the input of the model as (M T-N+1 ，M T-N+2 ，…，M T ) The output of the model is the multi-index vector predicted value of the T+1st predicted time
Figure SMS_11
S8, calculating the deviation degree of the multi-index vector predicted value according to the water environment monitoring threshold value of the ecologically sensitive area, and judging the water environment influence and the evaluation influence degree of the power transmission and transformation construction engineering, wherein the method specifically comprises the following steps:
s8-1, setting an ecologically sensitive area water environment monitoring threshold value gamma= { gamma according to the national surface water environment quality standard and the ecologically sensitive area historical water environment index l I l=1, …,5}, wherein
Figure SMS_12
Figure SMS_13
Is the first water environment index x l，t The upper boundary of the threshold value,γ l is the first water environment index x l，t Threshold lower bound.
S8-2, calculating a multi-index vector predicted value according to the water environment monitoring threshold value of the ecologically sensitive region
Figure SMS_14
Deviation epsilon = { epsilon% l I l=1,..2, 5}, wherein +.>
Figure SMS_15
Figure SMS_16
Is the deviation degree of the index of the first water environment,ε l the deviation degree is the first water environment index. />
Figure SMS_17
Andε l calculated by the following formula:
Figure SMS_18
Figure SMS_19
wherein,,
Figure SMS_20
for multi-index vector predictor->
Figure SMS_21
The first water environment index x l，t Predicted value at the t+1st predicted time.
And S8-3, judging the influence and degree of the water environment of the power transmission and transformation construction engineering according to the deviation epsilon of the multi-index vector predicted value.
(1) If it is
Figure SMS_22
And is also provided withε l Not less than 0,l E (1.. The first place, 5), the power transmission and transformation construction engineering affects the upper bound of the water environment threshold value, and shadows are generatedThe degree of ringing e is:
Figure SMS_23
(2) If it isε l < 0 and
Figure SMS_24
l epsilon (1,..5), the power transmission and transformation construction engineering affects the lower boundary of the water environment threshold, and the influence degree epsilon is:
∈＝{maxε l |ε l ＜0∩l∈[1，2，…，5]}
(3) If it is
Figure SMS_25
And is also provided withε l 0,l epsilon (1, 5), the power transmission and transformation construction engineering affects the upper and lower bounds of the water environment threshold, and the influence degree epsilon is as follows:
Figure SMS_26
(4) If it is
Figure SMS_27
And->
Figure SMS_28
l epsilon (1, 5), the power transmission and transformation construction engineering does not influence the water environment, and the influence degree epsilon is 0.
Compared with the prior art, the invention has the beneficial effects that:
according to the method for monitoring and evaluating the influence of the water environment of the power transmission and transformation construction project in the ecologically sensitive area, the rule of the evolution of the water environment index along with time is considered, a power transmission and transformation construction project water environment index prediction model based on a multi-index vector time sequence is constructed by a nonlinear mapping network with long-term trend and short-term fluctuation coordinated memory, compared with a traditional zero-dimensional, one-dimensional and two-dimensional water quality model, the model has stronger nonlinear mapping capability, the model parameter setting only depends on the multi-index vector historical time sequence and not on the experience of an evaluator, the difficulty of predicting the model parameter setting is solved, and the model has stronger adaptability; and secondly, the method replaces the absolute value judgment and evaluation of the water environment influence and the degree of the power transmission and transformation construction engineering water environment in the traditional method with the deviation degree of the multi-index vector predicted value based on the water environment monitoring threshold value of the ecological sensitive area, has the advantages of high judgment precision and simple and easy operation of the judgment mode, is more in line with the environment supervision practice of the ecological sensitive area, and can provide more accurate technical means for the water environment monitoring and protection of the power transmission and transformation construction engineering of the ecological sensitive area.
Drawings
FIG. 1 is a flow chart of a water environment monitoring and influence evaluating method for power transmission and transformation construction engineering in an ecologically sensitive area;
FIG. 2 is a water environment index prediction model diagram of a power transmission and transformation construction project based on a multi-index vector time sequence;
FIG. 3 shows the pH value of the multi-index vector predicted value of the water environment of the power transmission and transformation construction project in the invention;
FIG. 4 shows the COD of the water environment multi-index vector predicted value of the power transmission and transformation construction project in the invention;
FIG. 5 shows the predicted value of the multi-index vector of the water environment of the power transmission and transformation construction project in the invention 5 ；
FIG. 6 is a graph showing the NH content of ammonia nitrogen in the water environment multi-index vector predicted value of the power transmission and transformation construction project according to the present invention 3 -N；
FIG. 7 shows the petroleum type of the multi-index vector predicted value of the water environment of the power transmission and transformation construction engineering in the invention.
Detailed Description
The invention is described in further detail below with reference to the accompanying drawings and detailed description.
The principle of the invention is explained as follows:
the method is used for solving the water environment monitoring and protecting problems of power transmission and transformation construction engineering in the ecologically sensitive area. The invention is based on the following features: firstly, the water environment index of the ecologically sensitive area has regularity in evolution along with time, and secondly, the influence of power transmission and transformation construction engineering on the water environment can interfere and destroy the regularity. The method comprises the steps of firstly taking a sewage receiving body of a power transmission and transformation construction project as a monitoring object, constructing a water environment index prediction model based on a time sequence by using a nonlinear mapping network, constructing a learning sample according to a historical time sequence of multi-index vectors, training the water environment index nonlinear mapping network prediction model, identifying parameters of the model, finally obtaining a multi-index vector predicted value based on the prediction model, and judging the water environment influence and the degree of the power transmission and transformation construction project according to whether the multi-index vector predicted value exceeds an ecology sensitive area water environment monitoring threshold value. Compared with the traditional method, the method has the advantages of more accurate judging effect, concise judging mode and easy operation, and can provide technical support for the accurate monitoring and protection of the water environment of the power transmission and transformation construction engineering in the bio-sensitive area.
Examples:
in fig. 1, the influence of power transmission and transformation engineering on water environment mainly comprises two aspects: firstly, construction wastewater, including foundation construction slurry, construction site mixing wastewater, sand and stone material flushing and material cleaning and screening wastewater, machine repairing oil-containing wastewater and the like; and secondly, domestic sewage. Therefore, the following five water environment indexes of the sewage body of the power transmission and transformation construction project are focused on: pH value, chemical oxygen demand COD and five-day biochemical oxygen demand BOD 5 Ammonia nitrogen content NH 3 -N and petroleum. Meanwhile, the water environment indexes are easily influenced by various factors, so the hydrologic indexes of the power transmission and transformation construction engineering sewage receiving body in the embodiment focus on the following three types: water level, water temperature, flow rate.
In fig. 2, the model employs a hierarchical structure of input layer-hidden layer-output layer. In the input layer, the input of the nonlinear mapping network prediction model is the input of the s-th input-output mapping (M s ,M s+1 ,…,M s+N-1 ) The method comprises the steps of carrying out a first treatment on the surface of the In the output layer, the output of the model is the multi-index vector predicted value of the (s+N) th predicted time
Figure SMS_29
In the hidden layer, the node number of the hidden layer is equal to the input dimension N of the input-output mapping, and the nth node is marked as U n ，n∈[1,2,…,N]. For the nth node, four neurons f, i, g and o are adopted to realize the coordinated memory of long-term trend and short-term fluctuation of the time sequence, and the corresponding neuron outputs are as follows:
f n ＝σ(w f Z n +R f H n-1 +b f )
i n ＝σ(w i Z n +R i H n-1 +b i )
g n ＝δ(w g Z n +R g H n-1 +b g )
o n ＝σ(w o Z n +R o H n-1 +b o )
wherein f n ，i n ，g n And o n The excitation outputs of the four neurons f, i, g and o, σ (&) and δ (&) are sigmoid and tanh functions, b, respectively f ，b i ，b g And b o Deviations of four neurons, f, i, g and o, respectively, w f ，w i ， w g And w o Input weights of four neurons of f, i, g and o, respectively, R f ，R i ，R g And R is o Regression weights for the four neurons f, i, g and o, respectively. According to the model, the nth node U n Long-term memory C n And short-term memory H n The calculation is as follows:
Figure SMS_30
Figure SMS_31
wherein H is n-1 And C n-1 Characterizing the n-1 th node U respectively n-1 Short-term memory and long-term memory of (a). When n=n, then there are:
Figure SMS_32
this means the nth node U N Output H of (2) N Can be obtained by iterative calculation. H at this time N I.e. the multi-index vector predicted value of the s+N predicted time
Figure SMS_33
In the embodiment, under the condition of the technical support of sampling and detection, the sampling time interval is calculated in days, and the sampled water environment indexes are pH value, COD and BOD 5 、NH 3 -N and petroleum, the hydrographic environmental indicators of the samples are water level, water temperature, flow rate. The input dimension N of the input-output mapping takes a value of 50, the number of hidden layer nodes is 50, the number of training samples is 395, and the dimensions of the input weights and the regression weights are 200 x 8.
Fig. 3 to 7 show the deviation of the water environment multi-index vector predicted value of the power transmission and transformation construction engineering relative to the water environment monitoring threshold value of the ecologically sensitive area. In the figures, the ordinate respectively represents the pH value, the chemical oxygen demand COD and the five-day biochemical oxygen demand BOD of the multi-index vector predicted value of the water environment of the power transmission and transformation construction project 5 Ammonia nitrogen content NH 3 -N and petroleum. 5 times of iterative prediction are carried out on the multi-index vector of the water environment, and 396 to 400 sampling moments of the abscissa are corresponding. The water environment monitoring threshold is determined according to the upper and lower bounds of the historical water environment index of the ecological sensitive area reaching the standard. The upper and lower limits of pH are 8 and 7.5, respectively; the upper and lower limits of the chemical oxygen demand COD are 14.5mg/L and 11mg/L respectively; five days biochemical oxygen demand BOD 5 Upper and lower bounds of (2) and 1.5mg/L, respectively; ammonia nitrogen content NH 3 The upper and lower bounds of N are 0.5mg/L and 0.3mg/L, respectively; the upper and lower boundaries of the petroleum are 0.03mg/L and 0mg/L, respectively. The five water environment multi-index vector predicted values are all within the upper bound and the lower bound of the water environment monitoring threshold, which indicates that the power transmission and transformation construction engineering does not influence the water environment, and the influence degree epsilon is 0.

Claims (3)

1. The method for monitoring and evaluating the water environment and influence of the power transmission and transformation construction project in the ecological sensitive area is characterized by comprising the following steps of:
s1, atThe monitoring section and sampling point obtain the water environment index of the power transmission and transformation construction engineering sewage receiving body, including pH value, COD and BOD 5 、NH 3 -N and petroleum, constituting a temporal sequence of water environmental indicators: x is X t ＝{x 1，t ，x 2，t ，x 3，t ，x 4，t ，x 5，t }＝{x l，t |l＝[1，2，…，5]T is the sampling time, t.epsilon.1, 2, …, T]T represents the maximum sampling time, l is the first water environment index x l，t Number x of (x) 1，t The pH value is represented, and the method is dimensionless; x is x 2，t Represents chemical oxygen demand, COD, in milligrams per liter; x is x 3，t Representing the BOD of five days 5 Units are milligrams per liter; x is x 4，t Represents ammonia nitrogen content NH3-N, and the unit is milligrams per liter; x is x 5，t Indicating petroleum in milligrams per liter;
s2, acquiring hydrologic indexes of a sewage receiving body of the power transmission and transformation construction project at the monitoring section and the sampling point, wherein the hydrologic indexes comprise water level, water temperature and flow velocity, and a hydrologic index time sequence is formed: y is Y t ＝{H t ，C t ，V t T is the sampling time, t.epsilon.1, 2, …, T]，H t Representing the water level in meters; c (C) t Temperature is expressed in degrees celsius; v (V) t Indicating the water flow speed in meters per second;
s3, constructing a multi-index vector historical time sequence of the power transmission and transformation construction engineering sewage body through the water environment index time sequence and the hydrologic index time sequence:
M＝{M t |t∈[1，2，…，T]}，
wherein M is t ＝{X′ t ，Y t }，X′ t Is X t Is a subset of (i.e.)
Figure FDA0004084072480000011
X′ t The pH value, COD and BOD of the factors are evaluated by the influence of water environment 5 、NH 3 The number of the specific indexes is comprehensively determined according to the technical conditions of acquisition and monitoring and the environmental protection supervision requirements of the ecology sensitive area;
s4, according to the multi-index vector historical time sequence M= { M t |t∈[1，2，…，T]Building an input-output map; let N be the input dimension of the input-output map and N < T, then the multi-index vector historical time series M can construct T-N input-output maps:
(M 1 ，M 2 ，…，M N )→M N+1 ，
(M 2 ，M 3 ，…，M N+1 )→M N+2 ，
…
(M T-N ，M T-N+1 ，…，M T-1 )→M T ，
thus, the above-described input-output mapping is as follows:
(M s ，M s+1 ，…，M s+N-1 )→M s+N ，s∈[1，2，…，T-N]，
s represents the number of the s-th input-output mapping, and the value is [1,2, …, T-N]An integer therebetween; (M) s ，M s+1 ，…，M s+N-1 ) Representing the input of the s-th input-output mapping, M s+N Representing the expected output of the s-th input-output mapping;
s5, constructing a power transmission and transformation construction engineering water environment index nonlinear mapping network prediction model based on a multi-index vector historical time sequence, wherein the input of the model is the input of an input-output mapping, and the output is a multi-index vector predicted value at a predicted moment;
the nonlinear mapping network prediction model adopts a layered structure of an input layer, a hidden layer and an output layer: in the input layer, the input of the nonlinear mapping network prediction model is the input of the s-th input-output mapping (M s ，M s+1 ，…，M s+N-1 ) The method comprises the steps of carrying out a first treatment on the surface of the In the output layer, the output of the model is the multi-index vector predicted value of the (s+N) th predicted time
Figure FDA0004084072480000021
In the hidden layer, the node number of the hidden layer is equal to the input dimension N of the input-output mapping, and the nth node is marked as U n ，n∈[1，2，…，N]The method comprises the steps of carrying out a first treatment on the surface of the For the nth node, four neurons f, i, g and o are adopted to realize the coordinated memory of long-term trend and short-term fluctuation of the time sequence, and the corresponding neuron outputs are as follows:
f n ＝σ(w f Z n +R f H n-1 +b f )，
i n ＝σ(w i Z n +R i H n-1 +b i )，
g n ＝δ(w g Z n +R g H n-1 +b g )，
o n ＝σ(w o Z n +R o H n-1 +b o )，
wherein f n ，i n ，g n And o n The excitation outputs of the four neurons f, i, g and o, σ (&) and δ (&) are sigmoid and tanh functions, b, respectively f ，b i ，b g And b o Deviations of four neurons, f, i, g and o, respectively, w f ，w i ，w g And w o Input weights of four neurons of f, i, g and o, respectively, R f ，R i ，R g And R is o Regression weights for four neurons f, i, g and o, respectively; according to the model, the nth node U n Long-term memory C n And short-term memory H n The calculation is as follows:
Figure FDA0004084072480000022
Figure FDA0004084072480000023
wherein the method comprises the steps of
Figure FDA0004084072480000024
Representing multiplication of two matrices, H n-1 And C n-1 Characterizing the n-1 th node U respectively n-1 Short term memory and of (2)Long-term memory; when n=n, then there are:
Figure FDA0004084072480000025
this means the nth node U N Output H of (2) N Can be obtained by iterative calculation, in which case H N I.e. the multi-index vector predicted value of the s+N predicted time
Figure FDA0004084072480000026
S6, training a nonlinear mapping network prediction model of the water environment index of the power transmission and transformation construction project according to T-N input-output mappings, and identifying parameters of the model; the method comprises the following steps:
in S4, according to the multi-index vector history time sequence M= { M t |t∈[1，2，…，T]T-N input-output maps are constructed:
(M s ，M s+1 ，…，M s+N-1 )→M s+N ，s∈[1，2，…，T-N]，
s5, according to the power transmission and transformation construction engineering water environment index nonlinear mapping network prediction model, setting the input of the model as the input of the input-output mapping (M s ，M s+1 ，…，M s+N-1 ) The output of the model is the multi-index vector predicted value of the (s+N) th predicted time
Figure FDA0004084072480000027
The parameters of the model to be identified are W:
W＝{W i ，W r ，b}，
W i ＝{w f ，w i ，w g ，w o }，
W r ＝{R f ，R i ，R g ，R o }，
b＝{b f ，b i ，b g ，b o }，
wherein the method comprises the steps of，W i Input weights w representing four neurons f, i, g and o f ，w i ，w g And w o Is a collection of (3); w (W) r Regression weights R representing four neurons f, i, g and o f ，R i ，R g And R is o B represents the bias b of four neurons f, i, g and o f ，b i ，b g And b o Is a collection of (3);
the identification process of the model parameter W is as follows:
first calculate the training error E of the model train ：
Figure FDA0004084072480000031
The loss function J (W) of the model is determined as:
Figure FDA0004084072480000032
alpha is an adjustment factor for optimizing training error E in preference during iterative calculation train A trade-off is made between (a first term to the right of the equal sign of the J (W) expression) and a priority optimization model parameter W (a second term to the right of the equal sign of the J (W) expression);
the model parameters W are then determined using iterative calculations:
Figure FDA0004084072480000033
beta is the learning rate in iterative computation, and the change of W in each iterative computation is ensured to be in a proper range;
s7, predicting a multi-index vector predicted value at the moment T+1 by adopting a power transmission and transformation construction engineering water environment index nonlinear mapping network prediction model according to the finally determined model parameter W in the S6: setting the input of the model as (M T-N+1 ，M T-N+2 ，…，M T ) The output of the model is the T+1st predictionMulti-index vector predictor for time of day
Figure FDA0004084072480000034
S8, calculating the deviation degree of the multi-index vector predicted value according to the water environment monitoring threshold value of the ecologically sensitive area, and judging the water environment influence and the evaluation influence degree of the power transmission and transformation construction engineering, wherein the method specifically comprises the following steps:
s8-1, setting an ecologically sensitive area water environment monitoring threshold value gamma= { gamma according to the national surface water environment quality standard and the ecologically sensitive area historical water environment index l I l=1, …,5}, wherein
Figure FDA0004084072480000035
Figure FDA0004084072480000036
Is the first water environment index x l，t The upper boundary of the threshold value,γ l is the first water environment index x l，t A threshold lower bound;
s8-2, calculating a multi-index vector predicted value according to the water environment monitoring threshold value of the ecologically sensitive region
Figure FDA0004084072480000037
Deviation epsilon = { epsilon% l I l=1,..2, 5}, wherein +.>
Figure FDA0004084072480000038
Figure FDA0004084072480000039
Is the deviation degree of the index of the first water environment,ε l the deviation degree is the first water environment index; />
Figure FDA00040840724800000310
Andε l calculated by the following formula:
Figure FDA00040840724800000311
Figure FDA00040840724800000312
wherein,,
Figure FDA00040840724800000313
for multi-index vector predictor->
Figure FDA00040840724800000314
The first water environment index x l，t Predicted values at the t+1st predicted time;
s8-3, judging the influence and degree of the water environment of the power transmission and transformation construction engineering according to the deviation epsilon of the multi-index vector predicted value:
(1) If it is
Figure FDA0004084072480000041
And is also provided withε l Not less than 0,l epsilon (1, 5), the power transmission and transformation construction engineering affects the upper bound of the water environment threshold, and the influence degree epsilon is as follows:
Figure FDA0004084072480000042
(2) If it isε l < 0 and
Figure FDA0004084072480000043
the power transmission and transformation construction engineering affects the lower boundary of the water environment threshold, and the influence degree epsilon is as follows:
∈＝{maxε l |ε l ＜0∩l∈[1，2，…，5]}，
(3) If it is
Figure FDA0004084072480000044
And is also provided withε l ＜0，l epsilon (1,..5), the power transmission and transformation construction engineering affects the upper bound and the lower bound of the water environment threshold, and the influence degree epsilon is as follows:
Figure FDA0004084072480000045
(4) If it is
Figure FDA0004084072480000046
And->
Figure FDA0004084072480000047
The power transmission and transformation construction engineering does not influence the water environment, and the influence degree epsilon is 0.
2. The method for monitoring and evaluating the influence of the water environment of the power transmission and transformation construction project in the ecology sensitive area according to claim 1 is characterized in that: in S5, the value of the adjustment factor alpha is 0.1.
3. The method for monitoring and evaluating the influence of the water environment of the power transmission and transformation construction project in the ecology sensitive area according to claim 1 is characterized in that: in S5, the learning rate β in the iterative calculation takes a value of 0.001.
CN202210996244.4A 2022-08-18 2022-08-18 Ecological sensitive area power transmission and transformation construction engineering water environment monitoring and influence assessment method Active CN115358463B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202210996244.4A CN115358463B (en) 2022-08-18 2022-08-18 Ecological sensitive area power transmission and transformation construction engineering water environment monitoring and influence assessment method

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202210996244.4A CN115358463B (en) 2022-08-18 2022-08-18 Ecological sensitive area power transmission and transformation construction engineering water environment monitoring and influence assessment method

Publications (2)

Publication Number Publication Date
CN115358463A CN115358463A (en) 2022-11-18
CN115358463B true CN115358463B (en) 2023-06-30

Family

ID=84002772

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202210996244.4A Active CN115358463B (en) 2022-08-18 2022-08-18 Ecological sensitive area power transmission and transformation construction engineering water environment monitoring and influence assessment method

Country Status (1)

Country Link
CN (1) CN115358463B (en)

Families Citing this family (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115792141B (en) * 2023-01-06 2023-05-02 广州兰泰仪器有限公司 Method and system for improving balance detection efficiency of water activity meter
CN117491586B (en) * 2024-01-03 2024-03-19 江门市澳华生物科技有限公司 Water quality detection method and system

Family Cites Families (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US10488556B2 (en) * 2014-03-28 2019-11-26 Northeastern University System for multivariate climate change forecasting with uncertainty quantification
CN109886608A (en) * 2019-03-28 2019-06-14 交通运输部水运科学研究所 A kind of waterway engineering eco-environmental impact and Evaluation of Countermeasures based on Analysis on Mechanism
CN111639748B (en) * 2020-05-15 2022-10-11 武汉大学 Watershed pollutant flux prediction method based on LSTM-BP space-time combination model
CN112651665B (en) * 2021-01-14 2022-12-20 浙江鸿程计算机系统有限公司 Surface water quality index prediction method and device based on graph neural network

Also Published As

Publication number Publication date
CN115358463A (en) 2022-11-18

Similar Documents

Publication Publication Date Title
CN115358463B (en) Ecological sensitive area power transmission and transformation construction engineering water environment monitoring and influence assessment method
CN104965971B (en) A kind of ammonia nitrogen concentration flexible measurement method based on fuzzy neural network
Heddam et al. Predicting effluent biochemical oxygen demand in a wastewater treatment plant using generalized regression neural network based approach: a comparative study
Chen et al. Assessing wastewater reclamation potential by neural network model
CN109492265B (en) Wastewater effluent index prediction method based on dynamic nonlinear PLS soft measurement method
Han et al. Hierarchical neural network modeling approach to predict sludge volume index of wastewater treatment process
CN103093092B (en) The accident source electricity method that river sudden pollutant COD pollutes
CN111524030A (en) Plain river network area water environment monitoring early warning and safety guarantee management system
CN105574326A (en) Self-organizing fuzzy neural network-based soft measurement method for effluent ammonia-nitrogen concentration
Zhang et al. Study on water quality prediction of urban reservoir by coupled CEEMDAN decomposition and LSTM neural network model
Elhatip et al. Evaluation of water quality parameters for the Mamasin dam in Aksaray City in the central Anatolian part of Turkey by means of artificial neural networks
CN203772781U (en) Characteristic variable-based sewage total phosphorus measuring device
CN114878772B (en) Detection early warning monitoring system applied to sewage treatment
Wang et al. A Fusion Water Quality Soft‐Sensing Method Based on WASP Model and Its Application in Water Eutrophication Evaluation
Shi et al. A cumulative-risk assessment method based on an artificial neural network model for the water environment
CN107665288A (en) A kind of water quality hard measurement Forecasting Methodology of COD
Wang et al. Prediction Model of Agricultural Non-point Source Water Pollution Based on Grey Correlation Method.
Mohadesi et al. Use of ANFIS/genetic algorithm and neural network to predict inorganic indicators of water quality
Yang et al. Water quality assessment of deep learning-improved comprehensive pollution index: a case study of Dagu River, Jiaozhou Bay, China
Bhardwaj et al. Water Quality Modeling of the River Ganga in the Northern Region of India Using the Artificial Neural Network Technique.
Zhuang et al. Application of water quality evaluation model based on gray correlation analysis and artificial neural network algorithm
CN115330070B (en) Power transmission and transformation water environment index prediction method based on multi-factor coupling
Xiaoqin et al. On soft sensor of chemical oxygen demand by SOM-RBF neural network
Chai et al. Interval prediction of effluent TP for wastewater treatment plants
Chi et al. Wastewater treatment sensor fault detection using RBF neural network with set membership estimation

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant