CN112365077A - Construction method of intelligent storage scheduling system for power grid defective materials - Google Patents

Construction method of intelligent storage scheduling system for power grid defective materials Download PDF

Info

Publication number
CN112365077A
CN112365077A CN202011315338.8A CN202011315338A CN112365077A CN 112365077 A CN112365077 A CN 112365077A CN 202011315338 A CN202011315338 A CN 202011315338A CN 112365077 A CN112365077 A CN 112365077A
Authority
CN
China
Prior art keywords
model
defect
construction method
data
steps
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.)
Granted
Application number
CN202011315338.8A
Other languages
Chinese (zh)
Other versions
CN112365077B (en
Inventor
俞虹
唐诚旋
蒋群群
陈钰伊
张秀
程文美
代洲
徐一蝶
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.)
Guizhou Power Grid Co Ltd
Original Assignee
Guizhou Power Grid Co Ltd
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 Guizhou Power Grid Co Ltd filed Critical Guizhou Power Grid Co Ltd
Priority to CN202011315338.8A priority Critical patent/CN112365077B/en
Publication of CN112365077A publication Critical patent/CN112365077A/en
Application granted granted Critical
Publication of CN112365077B publication Critical patent/CN112365077B/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
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N20/00—Machine learning
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N20/00—Machine learning
    • G06N20/20—Ensemble learning
    • 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
    • G06N3/084—Backpropagation, e.g. using gradient descent
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N5/00—Computing arrangements using knowledge-based models
    • G06N5/01—Dynamic search techniques; Heuristics; Dynamic trees; Branch-and-bound
    • 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
    • G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
    • G06Q10/063—Operations research, analysis or management
    • G06Q10/0631—Resource planning, allocation, distributing or scheduling for enterprises or organisations
    • 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
    • G06Q10/08—Logistics, e.g. warehousing, loading or distribution; Inventory or stock management
    • G06Q10/087—Inventory or stock management, e.g. order filling, procurement or balancing against orders
    • G06Q10/0875—Itemisation or classification of parts, supplies or services, e.g. bill of materials
    • 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
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
    • Y02A30/00—Adapting or protecting infrastructure or their operation
    • 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)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Human Resources & Organizations (AREA)
  • General Physics & Mathematics (AREA)
  • Economics (AREA)
  • Software Systems (AREA)
  • Strategic Management (AREA)
  • Mathematical Physics (AREA)
  • Data Mining & Analysis (AREA)
  • Computing Systems (AREA)
  • Artificial Intelligence (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Evolutionary Computation (AREA)
  • General Engineering & Computer Science (AREA)
  • Development Economics (AREA)
  • General Business, Economics & Management (AREA)
  • Marketing (AREA)
  • Operations Research (AREA)
  • Quality & Reliability (AREA)
  • Tourism & Hospitality (AREA)
  • Game Theory and Decision Science (AREA)
  • Medical Informatics (AREA)
  • Computational Linguistics (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Finance (AREA)
  • Accounting & Taxation (AREA)
  • Health & Medical Sciences (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Biomedical Technology (AREA)
  • Biophysics (AREA)
  • General Health & Medical Sciences (AREA)
  • Molecular Biology (AREA)
  • Educational Administration (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

The invention discloses a construction method of an intelligent storage scheduling system for defective materials of a power grid, which comprises the steps of constructing a defective material prediction model based on open-source and purchased meteorological data and historical defect data provided by a power grid system, and predicting the defective materials; fusing the predicted result, and constructing a reinforcement learning model by combining the storage state of the current materials; and calculating a warehousing material scheduling scheme by using the reinforcement learning model, and visually displaying the warehousing material scheduling scheme. The invention serially connects the defect material prediction based on meteorological data and historical defect data and the defect material scheduling based on reinforcement learning, and integrates the prediction and the scheduling into a JavaWeb system, thereby realizing an intelligent and easy-to-use system.

Description

Construction method of intelligent storage scheduling system for power grid defective materials
Technical Field
The invention relates to the technical field of power grids and reinforcement learning, in particular to a construction method of a power grid defect material intelligent warehousing and dispatching system.
Background
The defect material prediction is based on the idea of data analysis and mining in each region, and a sequence-to-sequence model (sequence-to-sequence model) is constructed for different requirements of each region by using an artificial intelligence and machine learning method, so that the time sequence is predicted; the material management system realizes the scheduling of power grid defect materials by inputting and managing defect data, and the main platform technologies comprise a material management system based on JavaWeb, a material management system based on Python and the like
At present, although the prediction of the defective goods and materials can realize more accurate prediction, the prediction is isolated and the warehousing and scheduling cost is not considered; the reinforcement learning overall considers the warehousing and scheduling costs, the predicted data are also included in the algorithm, but no system combines the warehousing and scheduling costs and integrates the warehousing and scheduling costs into one system; the system is not well integrated from the aspects of meteorological data and historical defect data to reinforcement learning intelligent scheduling and data management, and the system is not combined with the meteorological data and the historical defect data to form a convenient management system.
Disclosure of Invention
This section is for the purpose of summarizing some aspects of embodiments of the invention and to briefly introduce some preferred embodiments. In this section, as well as in the abstract and the title of the invention of this application, simplifications or omissions may be made to avoid obscuring the purpose of the section, the abstract and the title, and such simplifications or omissions are not intended to limit the scope of the invention.
The present invention has been made in view of the above-mentioned conventional problems.
Therefore, the invention provides a construction method of an intelligent storage scheduling system for power grid defective goods and materials, which can solve the problems of prediction isolation of defective goods and materials and integration of reinforcement learning intelligent scheduling data.
In order to solve the technical problems, the invention provides the following technical scheme: constructing a defect material prediction model based on open-source and purchased meteorological data and historical defect data provided by a power grid system, and predicting the defect material; fusing the predicted result, and constructing a reinforcement learning model by combining the storage state of the current materials; and calculating a warehousing material scheduling scheme by using the reinforcement learning model, and visually displaying the warehousing material scheduling scheme.
The invention relates to a preferable scheme of a construction method of a power grid defective material intelligent storage scheduling system, wherein the construction method comprises the following steps: constructing a defect material prediction model comprises preprocessing the meteorological data and the historical defect data; taking a linear regression model, a ridge regression model, an elastic network model, a gradient lifting tree, an XgBoost model and a BP neural network model as the defect material prediction model; defining an optimization function and optimizing the defect material prediction model.
The invention relates to a preferable scheme of a construction method of a power grid defective material intelligent storage scheduling system, wherein the construction method comprises the following steps: the preprocessing comprises the steps of counting effective data in the data, filling null data and carrying out one-hot coding on the type data.
The invention relates to a preferable scheme of a construction method of a power grid defective material intelligent storage scheduling system, wherein the construction method comprises the following steps: the optimization function includes the selection of the optimal function,
Figure BDA0002791175260000021
where K is the number of classification regression trees, Γ is the classification regression tree space of the model, fkFor the kth classification regression tree of the model,
Figure BDA0002791175260000022
representative is the predicted value of the defect material.
The invention relates to a preferable scheme of a construction method of a power grid defective material intelligent storage scheduling system, wherein the construction method comprises the following steps: the fusing the prediction results comprises averaging and rounding the prediction results of the linear regression model, the ridge regression model, the elastic network model, the gradient lifting tree, the XgBoost model and the BP neural network model, as follows:
Figure BDA0002791175260000023
wherein F is a defect material required by the current time period, FpFor the p model needing to be fused, m represents the number of the classes needing the storage materials, and round is rounding.
The invention relates to a preferable scheme of a construction method of a power grid defective material intelligent storage scheduling system, wherein the construction method comprises the following steps: the constructing of the reinforcement learning model comprises the following steps:
Figure BDA0002791175260000031
wherein, V (S)t) For the current storage state S of the materialtValue function of rtFor the gain at the current time, γ is the attenuation factor, V (S)t+1) Is in a state St+1Value function of Xi,jAnd Bi,jRespectively representing the ex-warehouse quantity and the purchase quantity l of the goods and materials j in the warehouse i at the current momentjTo occupy capacityAmount, μiIs the upper limit of capacity, Zi,jThe quantity of materials j in the warehouse i at the current moment is shown, and n shows that n warehousing material warehouses are arranged.
The invention relates to a preferable scheme of a construction method of a power grid defective material intelligent storage scheduling system, wherein the construction method comprises the following steps: the benefit of the current time instant includes,
Figure BDA0002791175260000032
wherein Q isi,jRepresents the demand, p, of material j in the warehouse iiAnd piTo lose revenue, cjAmount spent purchasing goods, symbol (x)-Is composed of
Figure BDA0002791175260000033
The invention relates to a preferable scheme of a construction method of a power grid defective material intelligent storage scheduling system, wherein the construction method comprises the following steps: said state StAnd St+1Including, the storage state S of the current materialtComprises the following steps:
St＝Z∈Rn×m
wherein Z is the quantity of the materials in the warehouse, Rn×mIs n x m dimensional real number space;
St+1＝Z-X+B
wherein S ist+1And for the storage state after scheduling, X is a scheduling scheme, and B is a purchasing scheme.
The invention relates to a preferable scheme of a construction method of a power grid defective material intelligent storage scheduling system, wherein the construction method comprises the following steps: the warehousing material scheduling scheme comprises the following inequality which is required to be met:
Figure BDA0002791175260000034
the invention relates to a preferable scheme of a construction method of a power grid defective material intelligent storage scheduling system, wherein the construction method comprises the following steps: the method for computing the warehouse material scheduling scheme comprises the following steps of adopting an epsilon greedy strategy to determine an Action:
Figure BDA0002791175260000041
wherein A ist+1For the storage material scheduling scheme, Vw(St+1) To be in strategy w and said St+1The following cost function.
The invention has the beneficial effects that: and defect material prediction based on meteorological data and historical defect data and defect material scheduling based on reinforcement learning are connected in series and integrated into a JavaWeb system, so that an intelligent and easy-to-use system is realized.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings needed to be used in the description of the embodiments will be briefly introduced below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art to obtain other drawings based on these drawings without inventive exercise. Wherein:
fig. 1 is a schematic flow chart of a method for constructing a power grid defective material intelligent warehousing scheduling system according to a first embodiment of the present invention;
fig. 2 is a schematic structural diagram of a defective material prediction model based on meteorological data and historical defect data according to a method for constructing a power grid defective material intelligent warehousing and dispatching system according to a first embodiment of the invention;
fig. 3 is a schematic diagram of a warehousing scheduling engine based on reinforcement learning according to a method for constructing an intelligent warehousing scheduling system for defective materials in a power grid according to a first embodiment of the present invention;
fig. 4 is a schematic diagram of an intelligent defective material warehousing scheduling management system according to a method for constructing a power grid defective material intelligent warehousing scheduling system according to a second embodiment of the present invention;
fig. 5 is a schematic diagram of an intelligent defective material warehousing scheduling management system according to a method for constructing a power grid defective material intelligent warehousing scheduling system according to a second embodiment of the present invention;
fig. 6 is a schematic diagram of an intelligent defective material warehousing scheduling management system according to a method for constructing a power grid defective material intelligent warehousing scheduling system according to a second embodiment of the present invention;
fig. 7 is a schematic diagram of an intelligent defective material warehousing scheduling management system according to a method for constructing a power grid defective material intelligent warehousing scheduling system according to a second embodiment of the present invention.
Detailed Description
In order to make the aforementioned objects, features and advantages of the present invention comprehensible, specific embodiments accompanied with figures are described in detail below, and it is apparent that the described embodiments are a part of the embodiments of the present invention, not all of the embodiments. All other embodiments, which can be obtained by a person skilled in the art without making creative efforts based on the embodiments of the present invention, shall fall within the protection scope of the present invention.
In the following description, numerous specific details are set forth in order to provide a thorough understanding of the present invention, but the present invention may be practiced in other ways than those specifically described and will be readily apparent to those of ordinary skill in the art without departing from the spirit of the present invention, and therefore the present invention is not limited to the specific embodiments disclosed below.
Furthermore, reference herein to "one embodiment" or "an embodiment" means that a particular feature, structure, or characteristic described in connection with the embodiment is included in at least one implementation of the invention. The appearances of the phrase "in one embodiment" in various places in the specification are not necessarily all referring to the same embodiment, nor are separate or alternative embodiments mutually exclusive of other embodiments.
The present invention will be described in detail with reference to the drawings, wherein the cross-sectional views illustrating the structure of the device are not enlarged partially in general scale for convenience of illustration, and the drawings are only exemplary and should not be construed as limiting the scope of the present invention. In addition, the three-dimensional dimensions of length, width and depth should be included in the actual fabrication.
Meanwhile, in the description of the present invention, it should be noted that the terms "upper, lower, inner and outer" and the like indicate orientations or positional relationships based on the orientations or positional relationships shown in the drawings, and are only for convenience of describing the present invention and simplifying the description, but do not indicate or imply that the referred device or element must have a specific orientation, be constructed in a specific orientation and operate, and thus, cannot be construed as limiting the present invention. Furthermore, the terms first, second, or third are used for descriptive purposes only and are not to be construed as indicating or implying relative importance.
The terms "mounted, connected and connected" in the present invention are to be understood broadly, unless otherwise explicitly specified or limited, for example: can be fixedly connected, detachably connected or integrally connected; they may be mechanically, electrically, or directly connected, or indirectly connected through intervening media, or may be interconnected between two elements. The specific meanings of the above terms in the present invention can be understood in specific cases to those skilled in the art.
Example 1
Referring to fig. 1 to 3, a first embodiment of the present invention provides a method for constructing a power grid defective material intelligent warehousing scheduling system, including:
s1: and constructing a defect material prediction model based on open-source and purchased meteorological data and historical defect data provided by the power grid system, and predicting the defect materials.
The embodiment collects 15 areas in the jurisdiction or periphery of Guiyang city of Guizhou province: the method comprises the following steps of providing emergency material requirements provided by a power grid system of dolomite, north city, flower stream, Hui shui, Jinyang, Kaiyang, Longli, Nanming, Qingzhen, Shuanglong, Wudang, beacon, river, Huiwen and cloud rock, and historical meteorological data of a corresponding county, and preprocessing the data; specifically, valid data in the statistical data, such as a feature with a high null ratio removed and a feature with a high value ratio removed (for example, a feature with a value ratio exceeding 99% is a meaningless feature), is filled with null data, and the class type data is subjected to one-hot encoding.
Further, a linear regression model, a ridge regression model, an elastic network model, a gradient lifting tree, an XgBoost model and a BP neural network model are used as a defect material prediction model;
specifically, (1) the linear regression model is as follows:
Y＝a_1*X_1+a_2*X_2+a_3*X_3......a_n*X_n+b
where a _ n is a coefficient, X _ n is a variable, and b is a deviation.
(2) The ridge regression model is as follows:
min||Xw-y||2+z||w||2
wherein, X represents a characteristic variable, w represents a weight, y represents an actual value, and z is a square deviation factor.
(3) Elastic network model
min||Xw-y||2+z_1||w||+z_2||w||2
Where z _ n is a square deviation factor.
(4) The gradient boosting tree is as follows:
Figure BDA0002791175260000061
wherein, cmjIs the average number of pseudo-residuals, R, in the ith leaf node of the m-th treemjFor the leaf node region of the mth tree, M represents the number of iterations (i.e., the number of weak learners generated), J represents the number of leaf nodes per tree, x is data, and I is the median in the node sample.
(5) The XgBoost model is as follows:
Figure BDA0002791175260000062
wherein x isijData in dimensions i x j, theta is a parameter, yiIs the target variable.
(6) The BP neural network model is as follows:
yj＝(y1*W1)+(y2*W2)+...+(yi*Wi)+...+(yn*Wn)
wherein, XjIs the accumulation of i neurons, YiRepresents the amount of stimulation, W, delivered by a certain neuroniRepresenting the weight linking a certain neuron stimulus.
And further, defining an optimization function, and optimizing the defect material prediction model through the optimization function.
The optimization function is as follows,
Figure BDA0002791175260000071
where K is the number of classification regression trees, Γ is the classification regression tree space of the model, fkFor the kth classification regression tree of the model,
Figure BDA0002791175260000072
represented is the predicted defect material value, which is accumulated from the results of the K classification regression trees.
S2: and fusing the predicted result, and constructing a reinforcement learning model by combining the storage state of the current materials.
Averaging and rounding the prediction results of the linear regression model, the ridge regression model, the elastic network model, the gradient lifting tree, the XgBoost model and the BP neural network model, wherein the fusion result is as follows:
Figure BDA0002791175260000073
wherein F is a defect material required by the current time period, FpFor the p model needing to be fused, m represents the number of the classes needing the storage materials, and round is rounding.
Further, a reinforcement learning model is constructed, which is as follows:
Figure BDA0002791175260000074
wherein, V (S)t) For the current storage state S of the materialtValue function of rtFor the gain at the current time, γ is the attenuation factor, V (S)t+1) Is in a state St+1Value function of Xi,jAnd Bi,jRespectively representing the ex-warehouse quantity and the purchase quantity l of the goods and materials j in the warehouse i at the current momentjTo occupy capacity, muiIs the upper limit of capacity, Zi,jThe quantity of materials j in the warehouse i at the current moment is shown, and n shows that n warehousing material warehouses are arranged.
Specifically, the profit at the present time is as follows,
Figure BDA0002791175260000081
wherein Q isi,jRepresents the required quantity, p, of material j in warehouse iiAnd piLoss of revenue for local warehousing that can and cannot be scheduled without meeting demand, cjAmount spent purchasing goods, symbol (x)-Is composed of
Figure BDA0002791175260000082
Current storage state S of materialtComprises the following steps:
St＝Z∈Rn×m
wherein Z is the quantity of materials in the warehouse, Rn×mIs a real space of dimension n x m.
Scheduled warehouse status St+1Comprises the following steps:
St+1＝Z-X+B
wherein, X is a scheduling scheme, and B is a purchasing scheme.
Since the warehousing materials cannot be negative physically and the warehousing space is always limited, the warehousing material scheduling scheme (X, B) must satisfy the following inequality: the storage material scheduling scheme needs to satisfy the following inequality:
Figure BDA0002791175260000083
still further, the Bellman equation of the reinforcement learning model is changed into a form of data-driven online update, as follows:
V(St)←(1-αt)V(St)+αt[rt+γV(St+1)]
wherein alpha istThe learning rate at time t.
S3: and calculating a warehousing material scheduling scheme by using a reinforcement learning model, and visually displaying the warehousing material scheduling scheme.
Determining a warehousing material scheduling scheme Action by adopting an epsilon greedy strategy:
Figure BDA0002791175260000091
wherein A ist+1Scheduling schemes for storage materials, Vw(St+1) To be in policies w and St+1The following cost function.
Further, the steps of implementing visualization are as follows:
(1) firstly deploying software such as Tomcat, Nginx, Python, Django, MySQL, Java virtual machine and the like on a CentOS system'
(2) Storing the state data of the current storage materials in MySQL, then deploying a storage material demand prediction model in a Django framework, and developing a module for storing a model result in MySQL;
(3) and deploying a defect material scheduling algorithm based on reinforcement learning in the Django framework, calculating a scheduling scheme by combining the prediction result of S2, and writing the result of the scheduling algorithm into MySQL.
(4) And (3) developing a warehouse material management system based on an open source framework JeecgBoot, reading results in the steps (2) and (3), and visually displaying the results in a browser access mode.
Example 2
In order to verify and explain the technical effects adopted in the method, the method is adopted to predict the defects of the materials and visually display the scheduling of the materials.
Taking the defect number of the bulk material of Udang county 2015-2019 as an example, the related data is input into the defect prediction model of the method for prediction, and the prediction result is shown in the following table.
Table one: wudang county 2015-2019.
Time of day Real defect Predicting defects Accuracy rate
201512 month 1 1 100％
2016 (3 months) year 1 1 100％
2016 (5 months) year 2 2 100％
2016 (9 months) year 1 1 100％
10 months in 2017 1 1 100％
6 months in 2017 1 1 100％
7 month of 2017 1 1 100％
11 months in 2018 1 1 100％
4 months in 2018 1 1 100％
7 month in 2018 12 9 75％
9 month of 2018 2 2 100％
7 month in 2019 4 3 75％
The above table shows that the method can realize accurate prediction on the defect materials, and the prediction is basically consistent with the real data, and other counties can also achieve the accuracy similar to the data.
The method integrates beneficial effects of a power grid defect material prediction method based on meteorological data and historical defect data and an intelligent material warehousing and scheduling system based on reinforcement learning, and develops a power grid defect material warehousing and scheduling management system based on JavaWeb technology, as shown in FIG. 4; in the system, 3 menus are designed according to the types of materials, namely emergency defective materials, conventional defective materials and demand prediction of the defective materials, which are respectively shown in fig. 5, 6 and 7, and the three pages respectively show the number of currently proposed warehouses, the actual number of the warehouses and the number to be dispatched.
It should be noted that the above-mentioned embodiments are only for illustrating the technical solutions of the present invention and not for limiting, and although the present invention has been described in detail with reference to the preferred embodiments, it should be understood by those skilled in the art that modifications or equivalent substitutions may be made on the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, which should be covered by the claims of the present invention.

Claims (10)

1. A construction method of a power grid defect material intelligent warehousing dispatching system is characterized by comprising the following steps: comprises the steps of (a) preparing a mixture of a plurality of raw materials,
constructing a defect material prediction model based on open-source and purchased meteorological data and historical defect data provided by a power grid system, and predicting the defect material;
fusing the predicted result, and constructing a reinforcement learning model by combining the storage state of the current materials;
and calculating a warehousing material scheduling scheme by using the reinforcement learning model, and visually displaying the warehousing material scheduling scheme.
2. The construction method of the grid defect material intelligent warehousing dispatching system as claimed in claim 1, wherein: the constructing of the defect material prediction model comprises the following steps,
preprocessing the meteorological data and the historical defect data;
taking a linear regression model, a ridge regression model, an elastic network model, a gradient lifting tree, an XgBoost model and a BP neural network model as the defect material prediction model;
defining an optimization function and optimizing the defect material prediction model.
3. The construction method of the grid defect material intelligent warehousing dispatching system as claimed in claim 2, characterized in that: the pre-treatment comprises the steps of,
and counting effective data in the data, filling null data, and performing one-hot coding on the type data.
4. The construction method of the grid defect material intelligent warehousing dispatching system as claimed in claim 2 or 3, wherein: the optimization function includes the selection of the optimal function,
Figure FDA0002791175250000011
where K is the number of classification regression trees, Γ is the classification regression tree space of the model, fkFor the kth classification regression tree of the model,
Figure FDA0002791175250000012
representative is the predicted value of the defect material.
5. The construction method of the grid defect material intelligent warehousing dispatching system as claimed in claim 1 or 3, characterized in that: the fused prediction result comprises a result of the fusion prediction,
averaging and rounding the prediction results of the linear regression model, the ridge regression model, the elastic network model, the gradient lifting tree, the XgBoost model and the BP neural network model, as follows:
Figure FDA0002791175250000013
wherein F is a defect material required by the current time period, FpFor the p model needing to be fused, m represents the number of the classes needing the storage materials, and round is rounding.
6. The construction method of the grid defect material intelligent warehousing dispatching system as claimed in claim 5, wherein: the building of the reinforcement learning model comprises the following steps,
the reinforcement learning model is as follows:
Figure FDA0002791175250000021
wherein, V (S)t) For the current storage state S of the materialtValue function of rtFor the gain at the current time, γ is the attenuation factor, V (S)t+1) Is in a state St+1Value function of Xi,jAnd Bi,jRespectively representing the ex-warehouse quantity and the purchase quantity l of the goods and materials j in the warehouse i at the current momentjTo occupy capacity, muiIs the upper limit of capacity, Zi,jThe quantity of materials j in the warehouse i at the current moment is shown, and n shows that n warehousing material warehouses are arranged.
7. The method for constructing the power grid defective material intelligent warehousing dispatching system as claimed in claim 6, wherein: the benefit of the current time instant includes,
Figure FDA0002791175250000022
wherein Q isi,jRepresents the demand, p, of material j in the warehouse iiAnd piTo lose revenue, cjAmount spent purchasing goods, symbol (x)-Is composed of
Figure FDA0002791175250000023
8. The construction method of the grid defect material intelligent warehousing dispatching system as claimed in claim 6 or 7, wherein: said state StAnd St+1Comprises the steps of (a) preparing a mixture of a plurality of raw materials,
the storage state S of the current materialtComprises the following steps:
St＝Z∈Rn×m
wherein Z is the quantity of the materials in the warehouse, Rn×mIs n x m dimensional real number space;
St+1＝Z-X+B
wherein S ist+1And for the storage state after scheduling, X is a scheduling scheme, and B is a purchasing scheme.
9. The method for constructing the power grid defective material intelligent warehousing dispatching system as claimed in claim 8, wherein: the storage material scheduling scheme comprises the following steps of,
the storage material scheduling scheme needs to satisfy the following inequality:
Figure FDA0002791175250000031
10. the construction method of the grid defect material intelligent warehousing dispatching system as claimed in claim 1 or 9, wherein: the computing warehouse material scheduling scheme comprises the steps of,
determining Action by adopting an epsilon greedy strategy:
Figure FDA0002791175250000032
wherein A ist+1For the storage material scheduling scheme, Vw(St+1) To be in strategy w and said St+1The following cost function.
CN202011315338.8A 2020-11-20 2020-11-20 Construction method of intelligent storage scheduling system for power grid defective materials Active CN112365077B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202011315338.8A CN112365077B (en) 2020-11-20 2020-11-20 Construction method of intelligent storage scheduling system for power grid defective materials

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202011315338.8A CN112365077B (en) 2020-11-20 2020-11-20 Construction method of intelligent storage scheduling system for power grid defective materials

Publications (2)

Publication Number Publication Date
CN112365077A true CN112365077A (en) 2021-02-12
CN112365077B CN112365077B (en) 2022-06-21

Family

ID=74533089

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202011315338.8A Active CN112365077B (en) 2020-11-20 2020-11-20 Construction method of intelligent storage scheduling system for power grid defective materials

Country Status (1)

Country Link
CN (1) CN112365077B (en)

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112801418A (en) * 2021-03-16 2021-05-14 贵州电网有限责任公司 Power grid defect material prediction method
CN113516280A (en) * 2021-04-28 2021-10-19 贵州电网有限责任公司 Optimization method for power grid equipment fault probability prediction based on big data
CN113792923A (en) * 2021-09-10 2021-12-14 云南电网有限责任公司大理供电局 Visual scheduling window method based on power failure sensitivity prediction model
CN117333114A (en) * 2023-11-30 2024-01-02 广州欧派集成家居有限公司 Household product intelligent sorting, storing and managing system and method

Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20130030907A1 (en) * 2011-07-28 2013-01-31 Cbs Interactive, Inc. Clustering offers for click-rate optimization
CN109347149A (en) * 2018-09-20 2019-02-15 国网河南省电力公司电力科学研究院 Micro-capacitance sensor energy storage dispatching method and device based on depth Q value network intensified learning
CN109361601A (en) * 2018-10-31 2019-02-19 浙江工商大学 A kind of SDN route planning method based on intensified learning
US20200019983A1 (en) * 2018-07-13 2020-01-16 Pubwise, LLLP Digital advertising platform with demand path optimization
CN111639815A (en) * 2020-06-02 2020-09-08 贵州电网有限责任公司 Method and system for predicting power grid defect materials through multi-model fusion
CN111726217A (en) * 2020-06-29 2020-09-29 中南大学 Deep reinforcement learning-based autonomous frequency selection method and system for broadband wireless communication
CN111898211A (en) * 2020-08-07 2020-11-06 吉林大学 Intelligent vehicle speed decision method based on deep reinforcement learning and simulation method thereof

Patent Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20130030907A1 (en) * 2011-07-28 2013-01-31 Cbs Interactive, Inc. Clustering offers for click-rate optimization
US20200019983A1 (en) * 2018-07-13 2020-01-16 Pubwise, LLLP Digital advertising platform with demand path optimization
CN109347149A (en) * 2018-09-20 2019-02-15 国网河南省电力公司电力科学研究院 Micro-capacitance sensor energy storage dispatching method and device based on depth Q value network intensified learning
CN109361601A (en) * 2018-10-31 2019-02-19 浙江工商大学 A kind of SDN route planning method based on intensified learning
CN111639815A (en) * 2020-06-02 2020-09-08 贵州电网有限责任公司 Method and system for predicting power grid defect materials through multi-model fusion
CN111726217A (en) * 2020-06-29 2020-09-29 中南大学 Deep reinforcement learning-based autonomous frequency selection method and system for broadband wireless communication
CN111898211A (en) * 2020-08-07 2020-11-06 吉林大学 Intelligent vehicle speed decision method based on deep reinforcement learning and simulation method thereof

Non-Patent Citations (4)

* Cited by examiner, † Cited by third party
Title
FULI QIAO 等: "Trustworthy Edge Storage Orchestration in Intelligent Transportation Systems Using Reinforcement Learning", 《IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS》 *
ZHAOHUI HU 等: "Network load balancing strategy based on supervised reinforcement learning with shaping rewards", 《2013 FOURTH INTERNATIONAL CONFERENCE ON INTELLIGENT CONTROL AND INFORMATION PROCESSING》 *
李明玉: "能量收集MQAM无线通信系统基于强化学习的资源分配", 《中国优秀博硕士学位论文全文数据库(硕士) 信息科技辑》 *
陈明智 等: "基于强化学习的多机器人仓储调度方法研究", 《现代电子技术》 *

Cited By (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112801418A (en) * 2021-03-16 2021-05-14 贵州电网有限责任公司 Power grid defect material prediction method
CN112801418B (en) * 2021-03-16 2023-10-10 贵州电网有限责任公司 Method for predicting defective materials of power grid
CN113516280A (en) * 2021-04-28 2021-10-19 贵州电网有限责任公司 Optimization method for power grid equipment fault probability prediction based on big data
CN113792923A (en) * 2021-09-10 2021-12-14 云南电网有限责任公司大理供电局 Visual scheduling window method based on power failure sensitivity prediction model
CN117333114A (en) * 2023-11-30 2024-01-02 广州欧派集成家居有限公司 Household product intelligent sorting, storing and managing system and method
CN117333114B (en) * 2023-11-30 2024-04-02 广州欧派集成家居有限公司 Household product intelligent sorting, storing and managing system and method

Also Published As

Publication number Publication date
CN112365077B (en) 2022-06-21

Similar Documents

Publication Publication Date Title
CN112365077B (en) Construction method of intelligent storage scheduling system for power grid defective materials
Aalaei et al. A robust optimization model for cellular manufacturing system into supply chain management
Mulvey et al. Robust optimization of large-scale systems
Bas The integrated framework for analysis of electricity supply chain using an integrated SWOT-fuzzy TOPSIS methodology combined with AHP: The case of Turkey
Rezaei-Malek et al. An interactive approach for designing a robust disaster relief logistics network with perishable commodities
Saaty Axiomatic foundation of the analytic hierarchy process
CN112700045B (en) Intelligent site selection system based on land reserve implementation monitoring model
Charnes et al. Robustly efficient parametric frontiers via multiplicative DEA for domestic and international operations of the Latin American airline industry
CN107123982A (en) A kind of distribution network reliability analysis method of economic benefit based on equipment alteration
CN109992923A (en) A kind of transmission line of electricity paths planning method stage by stage based on variable resolution cost surface
CN103577942A (en) Decision support system and method for coordinated development of water resources and economy
Nozari et al. A Comprehensive Strategic-Tactical Multi-Objective Sustainable Supply Chain Model with Human Resources Considerations
Cheraghi et al. A multiproject scheduling and resource management model in projects construction
Luo et al. Optimization models and solving approaches in relief distribution concerning victims’ satisfaction: A review
Arbel A university budget problem: A priority-based approach
CN112258039B (en) Intelligent scheduling method for defective materials of power system based on reinforcement learning
Jones et al. Train timetabling and destination selection in mining freight rail networks: A hybrid simulation methodology incorporating heuristics
Radzki et al. Periodic planning of UAVs' fleet mission with the uncertainty of travel parameters
Day A recursive programing model for nonstructural flood damage control
CN116663834A (en) Aviation passenger scheduling method, device, computer equipment and storage medium
Debnath et al. Collaborative modelling for goal-oriented scenario planning: A resilience planning case study in the context of greater Sydney
Safaei et al. A multi-objective model to allocate multiple facilities at proposed locations in the multi-floor organisation, using an improved genetic algorithm. Case study: Isfahan Governorate
Say et al. Economic and mathematical model of substantiation of organizational network elements
CN113792828A (en) Power grid load prediction method, system, equipment and medium based on deep learning
Kazaryan Bases of Elements of Multi-Criteria Analysis of Quality of Design Decisions

Legal Events

Date Code Title Description
CB03 Change of inventor or designer information
CB03 Change of inventor or designer information

Inventor after: Yu Hong

Inventor after: Tang Chengxuan

Inventor after: Jiang Qunqun

Inventor after: Chen Jueyi

Inventor after: Zhang Xiu

Inventor after: Cheng Wenmei

Inventor after: Dai Zhou

Inventor after: Xu Yidie

Inventor before: Yu Hong

Inventor before: Tang Chengxuan

Inventor before: Jiang Qunqun

Inventor before: Chen Yuyi

Inventor before: Zhang Xiu

Inventor before: Cheng Wenmei

Inventor before: Dai Zhou

Inventor before: Xu Yidie

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