Warehouse environment conditioning method and system based on industrial internet of things

Abstract

The present application relates to the technical field of environmental control, in particular to a warehouse environment adjusting method and system based on industrial internet of things. The method comprises: acquiring indoor temperature data, outdoor temperature data and state sequence constituted by opening or closing state of each time in each independent storage area in the warehouse within a set time sequence; predicting the predicted indoor temperature of the next time based on the indoor temperature data, outdoor temperature data and state sequence of the set time sequence; outputting the control signal of the next time by using fuzzy control algorithm based on the predicted indoor temperature. The scheme of the present application can timely and accurately adjust the warehouse environment.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of environmental control. More particularly, the present application relates to a warehouse environment regulation method and system based on industrial internet of things. BACKGROUND

[0002] At present, the regulation of the environment in the warehouse is crucial to guarantee the quality and safety of the stored goods. Especially for warehouses of dangerous chemicals, medicines, high-value goods and other warehouses with strict requirements on environmental parameters. For example, most vaccines need to be kept at a refrigeration temperature of 2-8℃. Therefore, in the warehouse where the above special goods exist, the control of environmental parameters not only affects the preservation effect of goods, but also relates to the overall safety of the warehouse.

[0003] The traditional warehouse environment regulation method mainly relies on manual inspection and simple sensor data collection at regular intervals. Although this method can achieve the monitoring and regulation of the environment in the warehouse to a certain extent, it has obvious problems of insufficient regulation and control. Therefore, the environmental control of the stored goods in the warehouse is particularly important.

[0004] In related technologies, such as the Chinese patent application file with the publication number CN119472896A and the invention name "warehouse environment regulation method, system, device and medium based on industrial internet of things", it discloses that based on the collected image data in the warehouse, the stacked materials are divided to obtain a plurality of stacking units in the warehouse, and whether the environment in the warehouse needs to be regulated is determined by analyzing the environmental compatibility of each stacking unit.

[0005] The above scheme only considers the environmental compatibility of the goods itself to achieve the regulation and control of the environment in the warehouse; however, it does not consider that when the environment control of the independent storage area in the warehouse is performed, it may also be affected by external events, such as the influence of the door opening or closing state, so that the temperature of the independent storage area cannot be accurately controlled in a timely and effective manner, resulting in a safety risk of the goods in the independent storage area. Therefore, how to timely and effectively monitor and regulate the environment in the warehouse is particularly important. SUMMARY

[0006] The purpose of the present application is to provide a warehouse environment regulation method and system based on industrial internet of things to solve the problem that the environment of the independent storage area in the warehouse cannot be accurately controlled in a timely and effective manner in the prior art. To this end, the present application provides a scheme in the following two aspects.

[0007] In the first aspect, the present application provides a warehouse environment regulation method based on industrial internet of things, comprising:

[0008] Obtain indoor temperature data, outdoor temperature data and state sequence of each independent storage area in the warehouse within a set time sequence, wherein the state sequence is composed of opening or closing door state at each time point;

[0009] Predict the predicted indoor temperature at the next time point based on the indoor temperature data, outdoor temperature data and state sequence within the set time sequence;

[0010] Output the control signal at the next time point by using the fuzzy control algorithm based on the predicted indoor temperature;

[0011] The rule weight of the fuzzy control algorithm is:

[0012] Obtain the historical data set of each independent storage area, wherein each data point in the historical data set includes historical influence degree, difference between indoor temperature and target temperature and rule weight; the historical influence degree represents the change of indoor temperature at any time point in history;

[0013] Cluster all data points in the historical data set based on the historical influence degree and the difference, to obtain a plurality of clusters; calculate the rule weight of each data point in each cluster under the corresponding control rule; the rule weight is negatively correlated with the regulation and control time under the corresponding control rule;

[0014] Determine the predicted temperature at the next time point and the cluster to which the influence degree belongs, and select the rule weight of each control rule in the cluster as the rule weight at the next time point.

[0015] In the above scheme, by analyzing the historical influence degree and temperature difference of each data point in the historical data set of each independent storage area, time points with similar historical influence degree and temperature difference can be classified into a cluster, and the rule weight of multiple control rules under the same similar environment can be obtained, so that the rule weight of the fuzzy control algorithm at the next time point can be determined for temperature control at the next time point. That is, the scheme of the present application can solve the dynamic influence of opening or closing door state on the temperature in the independent storage area, and accurately control the temperature in the independent storage area in time and effectively.

[0016] Optionally, the indoor temperature data is collected by a sensor arranged near the door of each independent storage area; and the outdoor temperature data is collected by a sensor arranged near the door of each independent storage area.

[0017] Optionally, the influence degree is:

[0018] ;

[0019] wherein, is the predicted indoor temperature at the time point corresponding to the end of the window, a first time point, an indoor temperature at a first time point, a window length, a door opening state proportion, the door opening state proportion being a proportion of a number of door opening states in the window length; wherein the door opening or closing state at the first time point is the same as a door opening or closing state at a previous time point adjacent to the first time point.

[0020] The above scheme provides a method for accurately calculating an influence degree.

[0021] Optionally, the plurality of clusters is obtained through a K-mean clustering algorithm.

[0022] Through the clustering method, data with similar environments are classified into a category, thereby providing a basis for subsequent acquisition of rule weights.

[0023] Optionally, the rule weight is a normalized value of an inverse of a regulation time under a corresponding control rule; the regulation time is a time interval from starting regulation to reaching a target temperature.

[0024] Through the speed of the regulation time, the regulation effect under different control rules in a similar environment can be evaluated, and then the rule weight of each control rule in the corresponding cluster is obtained.

[0025] Optionally, the output of the control signal at the next time point comprises:

[0026] The difference between the predicted indoor temperature and the target temperature is input into a fuzzy control algorithm, and the control signal at the next time point is output.

[0027] Optionally, the predicted indoor temperature is obtained by using a recurrent neural network model.

[0028] Through the recurrent neural network model, the prediction of the indoor temperature at the next time point in the independent storage area can be realized.

[0029] Optionally, a training process of the recurrent neural network model comprises:

[0030] obtaining a sample data set; wherein the sample data set comprises historical indoor temperature data, historical outdoor temperature data and a historical state sequence in a historical setting time sequence;

[0031] inputting the sample data set into the recurrent neural network model for training, calculating a difference by using a cross-entropy loss, adjusting parameters of the recurrent neural network model by using a gradient descent algorithm, adjusting parameters of the network model, until a value of the cross-entropy loss between an output prediction result and a true result is less than a difference threshold, and obtaining a trained temperature prediction model.

[0032] Optionally, the method further comprises: performing time alignment processing on the collected indoor temperature data and outdoor temperature data to obtain indoor temperature data corresponding to the indoor temperature data after a delay time; and the delay time used in the time alignment processing is a time difference between two parameters corresponding to a maximum cross-correlation coefficient.

[0033] The time alignment processing on the indoor and outdoor temperature data ensures the consistency and accuracy of the collected indoor and outdoor temperature data, and provides a more reliable data basis for subsequent temperature prediction and control, thereby improving the stability and reliability of the environmental regulation of the independent storage areas in the warehouse.

[0034] In a second aspect, a warehouse environmental regulation system based on an industrial Internet of Things comprises:

[0035] a processor;

[0036] a memory storing computer instructions for warehouse environmental regulation based on an industrial Internet of Things, which, when executed by the processor, causes the system to perform the warehouse environmental regulation method based on an industrial Internet of Things described above.

[0037] The present application has the following beneficial effects:

[0038] The scheme of the present application can comprehensively collect and analyze indoor and outdoor temperature data and door opening or closing state information of each independent storage area in the warehouse, fully consider the influence of various factors on the temperature of the independent storage area, and more accurately reflect the actual temperature change of the warehouse compared to the traditional temperature regulation method which only considers a single factor or a simple threshold, thereby achieving more reasonable temperature control. BRIEF DESCRIPTION OF DRAWINGS

[0039] Figure 1 A step flowchart of the warehouse environmental regulation method based on an industrial Internet of Things in the embodiment is schematically shown;

[0040] Figure 2 A structural block diagram of the warehouse environmental regulation system based on an industrial Internet of Things in the embodiment is schematically shown. DETAILED DESCRIPTION

[0041] The technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application.

[0042] Since different independent storage areas are easily affected by outdoor temperature and door opening or closing operations (for example, frequent opening of the door will cause outdoor hot air to enter the warehouse, affecting the indoor temperature), the present application is directed to temperature control of different independent storage areas in the warehouse.

[0043] As Figure 1As shown, the warehouse environment adjustment method based on the industrial Internet of Things in the embodiment comprises the following steps:

[0044] Step S1, respectively acquiring indoor temperature data, outdoor temperature data and state sequence of each independent storage area in the warehouse within a set time sequence.

[0045] In this embodiment, the indoor temperature data is collected by sensors deployed near the door of each independent storage area; the outdoor temperature data is collected by sensors deployed near the door of each independent storage area.

[0046] Among them, the set time sequence includes the current time and the time before the current time; at the same time, the data points composed of the indoor temperature and the outdoor temperature at each time are labeled, label 1 is the open door state, and label 0 is the closed door state, so that the state sequence of the open or closed door state at all times can be obtained.

[0047] Further, the collected indoor temperature data and outdoor temperature data are also time-aligned. Specifically, by calculating the cross-correlation coefficient of indoor and outdoor temperature data, the time difference of the two parameters corresponding to the maximum cross-correlation coefficient is found as the delay time. For example, in a certain delay range (such as t1 time to t1+T time, T is a set interval time), the cross-correlation coefficient is calculated in a sliding window manner, and the time difference of the two data corresponding to the maximum cross-correlation coefficient is determined as the delay time.

[0048] According to the determined delay time, the indoor temperature data and the outdoor temperature data are time-aligned to ensure that the indoor temperature data and the outdoor temperature data are accurately aligned in time, providing more accurate data for subsequent temperature prediction and control in the independent storage area.

[0049] Step S2, based on the indoor temperature data, outdoor temperature data and state sequence of the set time sequence, predicting the predicted indoor temperature at the next time.

[0050] Among them, the predicted indoor temperature adopts a recurrent neural network model. Specifically, the recurrent neural network model can be an LSTM model.

[0051] The training process of the recurrent neural network model is as follows:

[0052] First, a sample data set is obtained. The sample data set includes historical indoor temperature data, historical outdoor temperature data and historical state sequence in a plurality of historical set time sequences; wherein the label is the indoor temperature at the next time of the last time in each historical set time sequence.

[0053] The time span of the data in the historical setting time sequence in the sample data set should be long enough to cover the warehouse temperature changes in different seasons and different time periods. For example, the indoor temperature data, outdoor temperature data and corresponding historical state sequence in different time periods every day in the past year are collected to ensure that the sample data is representative.

[0054] Secondly, the sample data set is input into the recurrent neural network model for training, and the difference is calculated by using cross-entropy loss. The gradient descent algorithm is used to adjust the parameters of the recurrent neural network model to adjust the parameters of the network model until the cross-entropy loss value between the predicted result and the true result is less than the difference threshold, and a trained temperature prediction model is obtained.

[0055] The difference threshold in the above can be 0 or a value close to 0, such as 0.01. Through multiple experiments and adjustment of model parameters, the model achieves good prediction effect.

[0056] In the embodiment, after obtaining the trained recurrent neural network model, the indoor temperature data, outdoor temperature data and state sequence of the setting time sequence are input into the recurrent neural network model, and the indoor temperature and door opening or closing state of the next moment of the current moment are obtained, realizing the prediction of the temperature of the independent storage area in the warehouse.

[0057] The above can accurately predict the indoor temperature of each independent storage area at the next moment by comprehensively considering the indoor and outdoor temperature, state sequence (door opening and closing frequency and door opening or closing duration) and other factors, and using the recurrent neural network model. Compared with the traditional method, the temperature change trend is predicted in advance, and the risk of damage to goods due to abnormal temperature is reduced.

[0058] Step S3, based on the predicted indoor temperature, a fuzzy control algorithm is used to output a control signal of the next moment.

[0059] The fuzzy control algorithm mainly includes the following steps: (1) defining input and output variables; (2) fuzzification; (3) establishing a fuzzy rule base; (4) fuzzy reasoning; (5) defuzzification; and (6) outputting a control signal.

[0060] The fuzzification step mainly calculates the membership degree, usually using a triangular membership function. Fuzzy reasoning is to combine rules and rule weights, aggregate the activation strength of each rule according to the membership degree and fuzzy rules to form a fuzzy output set.

[0061] The rule weight is a weighting coefficient of the fuzzy rule, and the range is 0-1.

[0062] The rule weight of the fuzzy control algorithm in the embodiment is obtained as follows:

[0063] Firstly, a historical data set of each independent storage area is obtained, and each data point in the historical data set includes a historical influence degree, a difference between an indoor temperature and a target temperature, a door opening or closing state sequence, and a corresponding control rule.

[0064] The historical influence degree is :

[0065] ;

[0066] The indoor temperature at the first end of the window at the t1 moment is , the indoor temperature at the first end of the window at the t1 moment is , the window length is , and the door opening state proportion is The door opening state duration proportion is the proportion of the number of door opening states in the window length.

[0067] The door opening state duration proportion is the proportion of the number of door opening states in the window length.

[0068] The above-mentioned influence degree at the t1 moment can be determined by analyzing the temperature change at the t1 moment and the door opening state proportion in the window. That is, the greater the temperature change rate in the window, the greater the influence of the outdoor temperature on the indoor temperature, and the greater the influence degree. The greater the door opening state proportion, the greater the number of door opening states in the window, and the greater the influence of the outdoor temperature on the indoor temperature over time.

[0069] The control rule is a control strategy executed by a device in the independent storage area under the conditions of the historical influence degree and the difference between the indoor temperature and the target temperature.

[0070] Exemplarily, for the environmental regulation of any independent storage area as a cold storage in a warehouse, the fuzzy rule is shown in Table 1 when fuzzy control is performed.

[0071] Table 1

[0072]

[0073] Exemplarily, if the target temperature is 2℃, NL is a temperature much lower than the target temperature, such as a current temperature of -5℃, NS is a temperature slightly lower than the target temperature, such as a current temperature of -1℃, ZE is a temperature close to the target temperature, such as a current temperature of 1℃, PS is a temperature slightly higher than the target temperature, such as a current temperature of 3℃, and PL is a temperature much higher than the target temperature, such as a current temperature of 5℃.

[0074] ​It should be noted that different independent storage areas can set the fuzzy rule base according to actual conditions, and are not limited to the content of the above fuzzy rule base. The fuzzy rule base of the same independent storage area is the same at different times.

[0075] Secondly, based on the difference between the historical influence degree, the indoor temperature and the target temperature, all data points in the historical data set are clustered to obtain a plurality of clusters.

[0076] In this embodiment, based on the difference between the historical influence degree, the indoor temperature and the target temperature, K-means algorithm is used to cluster all data points in the historical data set to obtain a plurality of clusters.

[0077] Each cluster represents a group of similar environmental states.

[0078] Then, the predicted indoor temperature at the next time and the cluster to which the influence degree belongs are determined, and the rule weight corresponding to each control rule in the cluster is selected as the rule weight at the next time.

[0079] In this embodiment, the predicted indoor temperature at the next time and the cluster to which the influence degree belongs are determined by calculating the Euclidean distance between the difference between the influence degree and the predicted indoor temperature at the next time and the target temperature and the center of each cluster.

[0080] The smaller the Euclidean distance, the more the environmental conditions at the next time belong to the corresponding cluster. The influence degree represents the change of the indoor temperature, which includes the influence of the outdoor temperature on the indoor temperature and the change of the indoor temperature itself.

[0081] The influence degree of the predicted indoor temperature in this embodiment is positively correlated with the change rate of the indoor temperature and the door opening frequency.

[0082] In one embodiment, the influence degree is:

[0083] ;

[0084] wherein, is the predicted indoor temperature at the time corresponding to the end of the window, is the indoor temperature at the time corresponding to the beginning of the window, is the window length, is the door opening state proportion.

[0085] The door opening state proportion is the proportion of the number of door opening states in the window length. The window length can be 5 or 10 times.

[0086] The door opening or closing state at the moment of the predicted indoor temperature can be considered to be the same as the door opening or closing state at the previous moment adjacent to the moment, for example, if the state at the previous moment is the door opening state, then the state at the moment of the predicted indoor temperature is the door opening state. This is because the two adjacent moments are relatively close, and therefore the instantaneous change of the predicted indoor temperature is relatively slow in theory, whether the state is the door opening state or the door closing state, and therefore the state at the previous moment is taken as the state at the next moment in the embodiment.

[0087] It should be noted that since the same cluster contains multiple data points, the control rules corresponding to each data point can be the same or different. When different, the cluster includes multiple different control rules; when the same, the cluster includes one control rule. Then, after determining the predicted temperature at the next moment and the cluster to which the influence degree belongs, when there are multiple different control rules in the cluster, the control rule at the moment of the predicted indoor temperature includes all the control rules in the cluster, and the rule weight of each control rule can be the normalized value of the inverse of the regulation time under the corresponding control rule.

[0088] The regulation time is the time interval from the start of regulation to the target temperature. The smaller the time interval, the better the regulation effect of the corresponding control rule, and therefore the larger the rule weight of the control rule.

[0089] The above normalization can use the maximum and minimum value normalization method; or the sum of the inverses of the regulation times under all different control rules is 1 to obtain the corresponding rule weight.

[0090] When there is only one control rule in the cluster, the corresponding rule weight is directly set to 1 without considering the regulation effect.

[0091] After determining the rule weight corresponding to the predicted indoor temperature, the difference between the predicted indoor temperature and the target temperature is input into the fuzzy control algorithm in the embodiment to activate the fuzzy rules, and the membership degrees under the control rules are weighted and output using the selected rule weight to realize temperature control at the next moment.

[0092] Since the fuzzy control algorithm is prior art, it will not be described in detail here.

[0093] The scheme of the application can be dynamically adjusted according to the actual environmental changes to adapt to the complex and variable warehouse environment, and improve the precise control of the independent storage area of the sensitive goods.

[0094] The application also provides a warehouse environment regulation system based on industrial Internet of Things. Figure 2As shown, the system comprises a processor and a memory having computer program instructions stored therein, which when executed by the processor implement the above-mentioned warehouse environment regulation method based on industrial Internet of Things according to the present application.

[0095] In the present application, the aforementioned memory can be any tangible medium that contains or stores a program that can be used by or in connection with an instruction execution system, apparatus, or device. For example, the computer-readable storage medium can be any appropriate magnetic storage medium or magneto-optical storage medium, such as Resistive Random Access Memory (RRAM), Dynamic Random Access Memory (DRAM), Static Random-Access Memory (SRAM), Enhanced Dynamic Random Access Memory (EDRAM), High-Bandwidth Memory (HBM), Hybrid Memory Cube (HMC), etc., or any other medium that can be used to store the desired information and that can be accessed by an application, module, or both. Any such computer storage media can be part of the device or accessible or connectable thereto. Any application or module described in the present application can be implemented by computer-readable / executable instructions stored or otherwise held by such computer-readable media.

[0096] In the description of the present application, the meaning of "a plurality of" is at least two, such as two, three or more, etc., unless otherwise explicitly specifically limited.

[0097] Although the present application has been shown and described with respect to several embodiments thereof, it will be apparent that those skilled in the art will be able to make many modifications, changes and alterations to the present application without departing from the spirit and scope thereof.

Claims

1. A warehouse environment control method based on the Industrial Internet of Things, characterized in that, include: The system acquires indoor temperature data, outdoor temperature data, and state sequence consisting of door open or closed status for each independent storage area in the warehouse within a set time period. Based on indoor temperature data, outdoor temperature data, and state sequence at a set time, predict the indoor temperature at the next moment. Based on the predicted indoor temperature, a fuzzy control algorithm is used to output the control signal for the next moment. The rule weights of the fuzzy control algorithm are: Obtain historical datasets for each independent storage area. Each data point in the historical dataset includes historical impact and the difference between indoor temperature and target temperature. Historical impact represents the change in indoor temperature at any point in history. Based on historical impact and difference, all data points in the historical dataset are clustered to obtain multiple clusters; the rule weight of each data point in each cluster is calculated under the corresponding control rule; the rule weight is negatively correlated with the control time under the corresponding control rule. Determine the cluster to which the predicted temperature and influence level belong at the next moment, and select the rule weights under each control rule in the cluster as the rule weights for the next moment. Historical influence for: ;in, Let be the indoor temperature at time t1 corresponding to the end of the window. The first point corresponding to the beginning of the window Indoor temperature at any given time For window length, The percentage of the door is in the open state; the percentage of the door remains open is the percentage of the number of times the door is open within the window length. Impact for: ;in, The first one corresponding to the end of the window Predicting indoor temperature at any time The first point corresponding to the beginning of the window Indoor temperature at any given time The percentage of doors in the open state is the proportion of doors in the open state within the window length; the first... The open or closed state of a door at a given moment is the same as the open or closed state of the door at the adjacent moment in time.

2. The warehouse environment control method based on the Industrial Internet of Things according to claim 1, characterized in that, The indoor temperature data is collected by sensors deployed near the entrance of each independent storage area; the outdoor temperature data is collected by sensors deployed near the outside of each independent storage area.

3. The warehouse environment control method based on the Industrial Internet of Things according to claim 1, characterized in that, The multiple clusters were obtained using the K-means clustering algorithm.

4. The warehouse environment control method based on the Industrial Internet of Things according to claim 1, characterized in that, The rule weight is the normalized value of the reciprocal of the control time under the corresponding control rule; the control time is the time interval from the start of control to reaching the target temperature.

5. The warehouse environment control method based on the Industrial Internet of Things according to claim 1, characterized in that, The control signal for the next time step includes: The difference between the predicted indoor temperature and the target temperature is input into the fuzzy control algorithm, which outputs the control signal for the next moment.

6. The warehouse environment control method based on the Industrial Internet of Things according to claim 1, characterized in that, The predicted indoor temperature is obtained using a recurrent neural network model.

7. The warehouse environment control method based on the Industrial Internet of Things according to claim 6, characterized in that, The training process of the recurrent neural network model is as follows: Obtain the sample dataset; the sample dataset includes historical indoor temperature data, historical outdoor temperature data, and historical state sequences within multiple historical time periods. The sample dataset is input into the recurrent neural network model for training. The difference is calculated using cross-entropy loss, and the parameters of the recurrent neural network model are adjusted using the gradient descent algorithm until the cross-entropy loss between the output prediction result and the actual result is less than the difference threshold, thus obtaining the trained temperature prediction model.

8. The warehouse environment control method based on the Industrial Internet of Things according to claim 1, characterized in that, It also includes: performing time alignment processing on the collected indoor temperature data and outdoor temperature data to obtain indoor temperature data after a delay time corresponding to the indoor temperature data; the delay time used in the time alignment processing is the time difference between the two parameters corresponding to the maximum cross-correlation coefficient.

9. A warehouse environment control system based on the Industrial Internet of Things, characterized in that, include: processor; A memory storing computer instructions for warehouse environment regulation based on the Industrial Internet of Things (IIoT), which, when executed by the processor, cause the system to perform the warehouse environment regulation method based on the IIoT according to any one of claims 1-8.

Images