Intelligent construction data optimization storage method

Abstract

The present application relates to the technical field of data storage, and particularly relates to a smart construction data optimization storage method, wherein the residual storage duration, the data type and the first storage priority calculated by each group of construction data are input into a preset regression model to obtain a predicted access frequency, which is used to predict the future access demand of the construction data; the time and space correlation degree between the new and old construction data belonging to the process is analyzed, the storage priority of the old data is dynamically calculated in combination with the storage priority of the new data, the storage value of the old data is adjusted in real time along with the advancement of the new process, the associated data with reference value for the current construction is prevented from being mistakenly eliminated due to isolated evaluation of the old data, the predicted data heat is calculated in combination with the predicted access frequency and the storage priority of the construction data, the data in the preset storage is updated according to the predicted data heat, the storage resources are allocated on demand, and the utilization rate of the storage resources and the future utilization rate of the data stored in the preset storage are improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of data storage, in particular to a smart construction data optimization storage method. BACKGROUND

[0002] In the smart construction scene, construction data has the characteristics of multi-source heterogeneity, large life cycle difference and strong business correlation. The storage optimization of construction data needs to balance the conditions such as real-time access, storage cost and compliance requirements. In the prior art, the smart construction data storage method is mostly based on fixed storage period (such as dividing cold and hot data according to time threshold) or historical access frequency to statically allocate storage resources, ignoring the inherent value difference of data types and the influence of remaining storage time length on access demand. Moreover, construction data has strong process correlation, but the existing method does not establish a priority association mechanism between new data and old data, causing storage resource allocation to lag, resulting in low utilization of storage resources and stored data.

[0003] Therefore, in the smart construction scene, how to optimize the data storage method to improve the utilization of storage resources and stored data has become a problem to be solved. SUMMARY

[0004] In view of the above technical problems, the technical scheme adopted by the present application is a smart construction data optimization storage method, which comprises the following steps:

[0005] S1, when receiving a data storage request, according to the collection time and data type of each group of first construction data to be stored and each group of second construction data already stored in the preset storage, the remaining storage time length corresponding to each group of first construction data and each group of second construction data is obtained.

[0006] S2, the remaining storage time length, data type and first storage priority corresponding to the data type of each group of first construction data and each group of second construction data are respectively input into the trained preset regression model, and the predicted access frequency corresponding to each group of first construction data and each group of second construction data is obtained.

[0007] S3, according to the predicted access frequency corresponding to each group of first construction data and the second storage priority corresponding to the construction process to which it belongs, the predicted data heat corresponding to each group of first construction data is obtained.

[0008] S4, according to the construction process to which each group of first construction data belongs and the corresponding second storage priority, and the construction process to which each group of second construction data belongs, the third storage priority corresponding to each group of second construction data is obtained.

[0009] S5, obtaining the predicted data heat degree corresponding to each group of second construction data according to the predicted access frequency and the third storage priority corresponding to each group of second construction data.

[0010] S6, updating the construction data stored in the preset storage according to the predicted data heat degree corresponding to all first construction data and second construction data.

[0011] The present application has at least the following beneficial effects: by calculating the remaining storage duration of each group of construction data, the data storage period is accurately matched with the actual life cycle, providing a quantitative basis for subsequent storage optimization in the time dimension, and avoiding unnecessary long-term occupation of resources by data; by inputting the remaining storage duration, data type and first storage priority into the trained preset regression model, the predicted access frequency is obtained, which is used to predict the future access demand of construction data; by combining the predicted access frequency of the first construction data with the second storage priority of the process to which the second construction data belongs, the predicted data heat degree is calculated, which more accurately reflects the future access value of new data to ensure that high-frequency construction data of key processes can obtain storage resources preferentially; by analyzing the spatio-temporal correlation degree of the second construction data and the process to which the first construction data belongs, the third storage priority of the old data is dynamically calculated, so that the storage value of the old data is adjusted in real time with the advancement of new processes, and the associated data with reference value to the current construction is prevented from being mistakenly eliminated due to isolated evaluation of old data; by combining the predicted access frequency of the second construction data with the third storage priority to calculate the predicted data heat degree, the value evaluation of the old data is integrated with the correlation degree with the new process, so that the old data with supporting effect on the current construction can obtain a reasonable storage priority; by updating the data in the preset storage according to the predicted data heat degree, high-value data is retained or added to the preset storage, and low-value data is eliminated, so that the storage resources are allocated on demand, and the utilization rate of the storage resources and the future utilization rate of the data stored in the preset storage are improved. BRIEF DESCRIPTION OF DRAWINGS

[0012] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings needed in the embodiment description will be briefly introduced. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor.

[0013] Figure 1 A flowchart of a smart construction data optimization storage method provided by the embodiments of the present application. DETAILED DESCRIPTION

[0014] With reference to the accompanying drawings, the technical solutions in the embodiments of the present application will be clearly and completely described below, obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work belong to the scope of protection of the present application.

[0015] It should be noted that the terms "first", "second" and the like in the specification and claims of the present application and the above-described drawings are used to distinguish similar mixing water systems, and do not have to be used to describe a specific order or sequence. It can be understood that the above-described terms for distinguishing similar mixing water systems can be interchanged under appropriate circumstances, so that the present application can also be implemented in other embodiments in addition to the above-described illustrated embodiments or described embodiments. In addition, the terms "include" and "have" and any variations thereof are intended to cover non-exclusive inclusion, for example, a process, method, system, product or server including a series of steps or units does not have to be limited to those steps or units clearly listed, but can include other steps or units not clearly listed or inherent to these processes, methods, products or devices.

[0016] The embodiment provides a wisdom construction data optimization storage method, which comprises the following steps, as shown in the figure: Figure 1

[0017] S1, when receiving a data storage request, according to the collection time and data type of each group of first construction data to be stored and each group of second construction data already stored in the preset storage, the remaining storage time corresponding to each group of first construction data and each group of second construction data is obtained.

[0018] The collection time is the specific time when the data is generated, which is used to calculate the data storage time to judge the data timeliness.

[0019] The data type can be set by the implementer according to the actual situation, for example, the data type is divided according to the business attribute in the embodiment, including construction progress management data, construction quality management data, construction safety management data and construction cost management data, which is used to divide the inherent life cycle and basic storage priority of construction data.

[0020] Specifically, the construction progress management data can include plan data, actual progress data, deviation analysis data and the like, the construction quality management data can include material detection data, process acceptance data, defect rectification data and the like, the construction safety management data can include hidden danger investigation data, equipment state data, emergency disposal data and the like, and the construction cost management data can include budget data, visa data, payment voucher data and the like.

[0021] ​Different types of construction data have different inherent life cycles. For example, construction safety management data needs to be accessed frequently during construction, but the demand for access decreases sharply after completion. Construction quality management data needs to be saved for a long time to meet the compliance audit requirements.

[0022] The preset memory is used to dynamically adjust the stored construction data according to data heat. Since the storage space in the preset memory is limited, when a data storage request is received and the storage space is insufficient, the heat of the first construction data to be stored and the second construction data already stored needs to be measured to preferentially store construction data with higher heat and delete construction data with lower heat, thereby improving the rationality of storage resource allocation. Construction data can also be divided into hot storage area, warm storage area and cold storage area according to data heat, and construction data can be migrated and adjusted between different storage areas.

[0023] In a specific embodiment, S1 includes the following steps:

[0024] S11, each group of first construction data and each group of second construction data are taken as reference construction data.

[0025] S12, for any group of reference construction data, according to the data type corresponding to the current reference construction data, a plurality of groups of historical construction data corresponding to the current data type and the actual storage duration corresponding to each group of historical construction data are obtained.

[0026] S13, according to the actual storage duration corresponding to each group of historical construction data, the reference storage duration corresponding to the current data type is obtained.

[0027] S14, according to the collection time corresponding to each group of reference construction data and the current time, the stored duration corresponding to each group of reference construction data is obtained.

[0028] S15, according to the stored duration corresponding to each group of reference construction data and the corresponding reference storage duration, the remaining storage duration corresponding to each group of reference construction data is obtained.

[0029] Among them, the historical construction data can be the construction process record that has completed the life cycle and no longer needs to be frequently accessed. The time span from generation to the last access is saved completely as the actual storage duration, which is used to predict the future access value of similar data and guide the heat measurement of construction data.

[0030] According to the storage time point and the deletion time point of each set of historical construction data in the preset storage, the actual storage duration corresponding to each set of historical construction data is obtained, and according to the actual storage duration corresponding to each set of historical construction data, the reference storage duration of the current data type is obtained through a statistical method. For example, the median of the actual storage duration corresponding to all sets of historical construction data is taken as the reference storage duration, or the 90th percentile of the actual storage duration corresponding to all sets of historical construction data is taken as the reference storage duration, so as to avoid the influence of abnormal values.

[0031] The stored duration = current time - collection time, which is used to reflect the survival time of the reference construction data from collection to now.

[0032] The remaining storage duration = reference storage duration - stored duration, which is used to reflect the future access value of the reference construction data.

[0033] The above introduces a historical data learning mechanism, which no longer depends on a fixed basic storage period, but dynamically adjusts the reference storage duration by analyzing the actual storage behavior of similar data, so that the calculation of the remaining storage duration is closer to the real business demand.

[0034] S2, the remaining storage duration, data type and first storage priority corresponding to each set of first construction data and each set of second construction data are input into the trained preset regression model respectively, and the predicted access times corresponding to each set of first construction data and each set of second construction data are obtained.

[0035] Among them, the inherent attributes (data type, first storage priority) and time attributes (remaining storage duration) of the construction data jointly determine the number of times the construction data is accessed in the future, and the trained preset regression model captures the potential relationship between the remaining storage duration, data type and first storage priority and the access times through learning of historical construction data,

[0036] The first storage priority is a basic importance level preset based on the business value and compliance requirements of the data type, which is used to reflect the basic storage value of the construction data and affects the storage resources of the construction data in the preset storage, and can be input into the preset regression model to affect the prediction result, for example, the construction data with high first storage priority has more predicted access times.

[0037] In a specific embodiment, the specific values of the first storage priority corresponding to each data type can be set by the implementer according to actual conditions, for example, by the implementer or an expert in combination with the importance of construction business, compliance requirements, and access urgency. For example, the first storage priority corresponding to construction safety management data is 5, the first storage priority corresponding to construction quality management data is 4, the first storage priority corresponding to construction safety management data is 3, and the first storage priority corresponding to construction cost management data is 3.

[0038] The preset regression model is a pre-trained machine learning model, usually adopting an encoder and a fully connected layer architecture, for mapping input features to predicted access times of continuous values, representing the number of times the corresponding construction data is expected to be accessed within the remaining storage duration in the future, as a data basis for subsequent heat calculation. Wherein, any architecture of the regression model in the prior art falls within the protection scope of the present application, and will not be described here.

[0039] As described above, the preset regression model extracts and analyzes the remaining storage duration, data type, and first storage priority by multi-feature fusion, outputs the predicted access times representing the number of times the corresponding construction data is expected to be accessed within the remaining storage duration in the future, and provides a data basis for analyzing data heat.

[0040] In a specific embodiment, the intelligent construction data optimized storage method further includes the following steps:

[0041] S10, a plurality of groups of historical construction data and a plurality of accessed times corresponding to each group of historical construction data are obtained.

[0042] S20, for any group of historical construction data, a plurality of remaining storage durations corresponding to the current historical construction data and the number of accesses in each remaining storage duration are obtained according to the collection time corresponding to the current historical construction data and the plurality of accessed times, and the reference storage duration corresponding to the data type corresponding to the current historical construction data.

[0043] S30, each remaining storage duration corresponding to the current historical construction data, the data type, and the first storage priority corresponding to the data type are taken as a group of training samples, and the number of accesses corresponding to each remaining storage duration is taken as the training label of the corresponding training sample.

[0044] S40, all groups of historical construction data are traversed to obtain all training samples and training labels corresponding to each training sample.

[0045] S50, the initial regression model is trained according to all training samples and training labels corresponding to each training sample, and a trained preset regression model is obtained.

[0046] The initial regression model is input with each set of training samples, and a predicted access frequency sample corresponding to each set of training samples is obtained. The model loss is calculated according to all the predicted access frequency samples and corresponding training labels, the model parameters are updated by minimizing the model loss, and the trained preset regression model is obtained.

[0047] On the basis of obtaining the training samples and the training labels, those skilled in the art know that the training method of the regression model in the prior art falls within the protection scope of the present application, and will not be described here. For example, the model loss can be calculated by using a mean square error loss function.

[0048] The above describes that the supervised learning data set is constructed by using the historical construction data, the initial regression model is trained to predict the mapping relationship between the remaining storage duration, the data type, the first storage priority and the access frequency, and a data basis is provided for subsequent storage optimization.

[0049] In a specific embodiment, S20 includes the following steps:

[0050] S201, obtaining a reference storage time point corresponding to the current historical construction data according to the collection time corresponding to the current historical construction data and the reference storage duration.

[0051] S202, obtaining an ith reference time point corresponding to the current historical construction data according to the collection time corresponding to the current historical construction data and the ith preset step corresponding to the data type corresponding to the current historical construction data, wherein the total number N of the reference time points corresponding to the current historical construction data is T / Δt, T is the reference storage duration corresponding to the current historical construction data, Δt is the preset step corresponding to the data type corresponding to the current historical construction data, rounddown() is a rounding down function, and i=1, 2, …N.

[0052] S203, obtaining an ith remaining storage duration according to the ith reference time point corresponding to the current historical construction data and the reference storage time point.

[0053] S204, determining the access frequency corresponding to the current historical construction data in the ith remaining storage duration as the access frequency corresponding to the current historical construction data in the ith reference time point to the reference storage time point according to the plurality of accessed time points corresponding to the current historical construction data.

[0054] The reference storage time point = collection time + reference storage duration, which is used to determine the cutoff time point at which the data theoretically no longer needs to be accessed, as the end point of the time window division.

[0055] The ith reference time point = the collection time + i x At, and the ith remaining storage duration = the reference storage time point - the ith reference time point.

[0056] The ith time window is [the ith reference time point, the reference storage time point), and if a certain access time t satisfies the ith reference time point ≤ t < the reference storage time point, the access is counted in the ith time window, and the number of accesses corresponding to each remaining storage duration is obtained accordingly.

[0057] The preset step length is a time interval used to divide the time period from the reference time point to the reference storage time point into a plurality of continuous time intervals (each interval corresponds to a remaining storage duration). The specific value of the preset step length can be set by the implementer according to the actual situation. For example, the implementer or an expert sets the preset step length in combination with the access frequency, timeliness, and variation speed of the construction data. Specifically, the more frequent the access, the stronger the timeliness, and the faster the variation, the smaller the preset step length, so as to capture the access rule more finely, and vice versa. For example, if the unit of the reference storage duration is day, the preset step length corresponding to the construction safety management data is 1 day, the preset step length corresponding to the construction quality management data is 3 days, the preset step length corresponding to the construction safety management data is 2 days, and the preset step length corresponding to the construction cost management data is 7 days.

[0058] In the above, the continuous access time sequence of the historical construction data is converted into the number of accesses in the discrete time window, which is used as the training label of the regression model, thereby improving the training rationality of the initial regression model and the output accuracy of the trained preset regression model.

[0059] S3, according to the predicted number of accesses corresponding to each set of first construction data and the second storage priority corresponding to the construction process to which the first construction data belongs, obtains the predicted data heat corresponding to each set of first construction data.

[0060] In a specific embodiment, S3 includes the following steps:

[0061] S31, for any set of first construction data, the product of the predicted number of accesses corresponding to the current first construction data and the second storage priority corresponding to the construction process to which the first construction data belongs is determined as the predicted data heat corresponding to the current first construction data.

[0062] S32, all sets of first construction data are traversed to obtain the predicted data heat corresponding to each set of first construction data.

[0063] The construction procedure is a series of ordered operation steps of the building project from the start to the completion. The specific classification form of the construction procedure can be set by the implementer according to the actual situation, for example, according to the construction logic of the building entity from bottom to top and from inside to outside, the construction procedure is divided into foundation and foundation engineering procedure, main structure engineering procedure, decoration engineering procedure, roof engineering procedure and equipment installation engineering procedure.

[0064] Specifically, the foundation and foundation engineering procedure includes site leveling, earthwork excavation, pile foundation construction, foundation treatment, underground waterproof engineering and the like. The main structure engineering procedure includes wall building, structural column pouring, steel bar binding, concrete pouring, steel member processing, anti-corrosion coating and the like. The decoration engineering procedure includes facing brick sticking, finishing, curtain wall installation, door and window frame installation and the like. The roof engineering procedure includes roof base treatment, waterproof construction, insulation layer laying and the like. The equipment installation engineering procedure includes water supply and drainage engineering, electrical engineering, heating and ventilation air conditioning engineering and the like.

[0065] The second storage priority is a dynamic importance level set according to the criticality of the construction procedure, which reflects the relative importance of the first construction data in a specific construction stage and the expected heat of the future access of the first construction data.

[0066] The specific value of the second storage priority corresponding to each construction procedure can be set by the implementer according to the actual situation, for example, the implementer or expert sets according to the criticality, risk level and data access frequency of the construction procedure, combined with the influence degree of the construction procedure on the overall quality, safety and progress of the project. For example, the second storage priority corresponding to the foundation and foundation engineering procedure is 5, the second storage priority corresponding to the main structure engineering procedure is 5, the second storage priority corresponding to the decoration engineering procedure is 4, the second storage priority corresponding to the roof engineering procedure is 3, and the second storage priority corresponding to the equipment installation engineering procedure is 2.

[0067] The access heat of the construction data is determined by the access frequency and the business importance, and the data value of the high access frequency and high priority procedure is highlighted through the product operation.

[0068] The above combines the predicted access frequency with the importance of the construction procedure of the first construction data to calculate the predicted data heat, which provides a ranking basis for the subsequent allocation of storage resources.

[0069] S4, according to the construction procedure and the corresponding second storage priority of each group of first construction data, and the construction procedure of each group of second construction data, the third storage priority corresponding to each group of second construction data is obtained.

[0070] The construction of the new process often relies on the data of the historical similar or related processes. According to the correlation between the construction process to which the second construction data belongs and the construction process to which the first construction data belongs, the storage priority of the old data is dynamically adjusted, so that the high priority of the new data is transmitted to the related old data through the space-time correlation degree, and the key old data is ensured not to be mistakenly eliminated, thereby improving the effectiveness and utilization rate of the stored data.

[0071] In an embodiment, S4 comprises the following steps:

[0072] S41, acquiring the time parameter and the space parameter corresponding to each construction process.

[0073] S42, for any group of second construction data, acquiring the time correlation degree between the current second construction data and each group of first construction data according to the time parameter corresponding to the construction process to which the current second construction data belongs and the time parameter corresponding to the construction process to which each group of first construction data belongs.

[0074] S43, acquiring the space correlation degree between the current second construction data and each group of first construction data according to the space parameter corresponding to the construction process to which the current second construction data belongs and the space parameter corresponding to the construction process to which each group of first construction data belongs.

[0075] S44, acquiring the third storage priority corresponding to each group of second construction data according to the time correlation degree and the space correlation degree between the current second construction data and each group of first construction data and the second storage priority corresponding to each group of first construction data.

[0076] The time parameter is a time interval describing the construction of the process, including the start time of the process and the end time of the process.

[0077] The space parameter is a physical range describing the construction of the process, including the construction area coordinates, floor number and building number.

[0078] The time correlation degree is used to measure the degree of connection or overlap in time between the process to which the old second construction data belongs and the process to which the new first construction data belongs, and the value range is [0, 1]. Wherein, 1 represents complete correlation in time, and 0 represents no correlation in time. For example, the time interval of the construction process corresponding to the first construction data is [T11, T12], and the time interval of the construction process corresponding to the second construction data is [T21, T22], then the time correlation degree G1 between the two construction data is as follows:

[0079] G1=1-|(T11+T12) / 2-(T21+T22) / 2| / (max(T12, T22)-min(T11, T21)), wherein max() is a maximum function, min() is a minimum function,

[0080] The spatial correlation degree is used to measure the overlapping or proximity degree of the old second construction data belonging to the process and the new first construction data belonging to the process in the construction position, and the value range is [0, 1]. Wherein, 1 represents the same position, and 0 represents no spatial correlation. Then the spatial correlation degree G2 between the second construction data and the first construction data is S1 / S2, wherein S1 is the overlapping area of the construction area of the corresponding second construction data belonging to the construction process and the corresponding first construction data belonging to the construction process, and S2 is the average area of the construction area of the corresponding second construction data belonging to the construction process and the corresponding first construction data belonging to the construction process.

[0081] For any group of second construction data and any group of first construction data, the product of the time correlation degree between the current second construction data and the current first construction data, the spatial correlation degree and the second storage priority corresponding to the current first construction data is determined as the reference priority of the current second construction data for the current first construction data.

[0082] Further, the sum of the reference priorities of the current second construction data for all first construction data is determined as the third storage priority corresponding to the current second construction data.

[0083] The above, by measuring the correlation degree of the old second construction data belonging to the process and the new first construction data belonging to the process in time and space, the second storage priority of the first construction data is combined to update the third storage priority of the second construction data, so as to dynamically adjust the priority of the old data with the advancement of the construction process, realize the value linkage of new and old construction data, and make the storage strategy more suitable for the continuity and correlation demand of construction business.

[0084] S5, according to the predicted access frequency and the third storage priority corresponding to each group of second construction data, obtains the predicted data heat corresponding to each group of second construction data.

[0085] In a specific embodiment, S5 includes the following steps:

[0086] S51, for any group of second construction data, the product of the predicted access frequency corresponding to the current second construction data and the third storage priority is determined as the predicted data heat corresponding to the current second construction data.

[0087] S52, traversing all groups of second construction data, obtaining the predicted data heat corresponding to each group of second construction data.

[0088] The third storage priority is the importance level of the second construction data dynamically adjusted with the advancement of the new construction process, and is used to reflect the expected heat of the future access of the second construction data.

[0089] The access frequency of the construction data is determined by the access times and the importance of the business, and the product operation is used to realize the nonlinear fusion, so as to highlight the data value of the high access times and high priority process.

[0090] The predicted access times are combined with the importance of the second construction data which dynamically changes with the progress of the construction process to calculate the predicted data frequency, thereby providing a sorting basis for the subsequent allocation of storage resources.

[0091] S6, updating the stored construction data in the preset storage according to the predicted data frequency of all the first construction data and the second construction data.

[0092] In an embodiment, S6 includes the following steps:

[0093] S61, sorting all the first construction data and the second construction data in descending order of the predicted data frequency to obtain a data sorting result.

[0094] S62, obtaining the storable construction data according to the data amount of each group of first construction data and each group of second construction data, and the storage space of the preset storage.

[0095] S63, updating the storable construction data as the stored construction data in the preset storage.

[0096] The predicted data frequency reflects the future importance of the corresponding construction data, and the higher the predicted frequency, the more the data should be retained, so in the limited storage space, the data is divided into storable data and data to be eliminated by sorting and data amount calculation, the storable data is stored or kept in the preset storage, and the data to be eliminated is deleted.

[0097] Specifically, the total data amount is initialized to 0, the storable data list is an empty set, the sorted data list is traversed in turn, the traversed construction data is added to the storage data list, and the data amount is correspondingly accumulated in turn, if the accumulated data amount exceeds the remaining space of the preset storage, the traversal is stopped, the current storable data list is returned, and the construction data after the last storable data is sorted is taken as the data to be eliminated.

[0098] The above, by calculating the remaining storage duration of each set of construction data, the data storage period is accurately matched with the actual life cycle, providing a quantitative basis for time dimension for subsequent storage optimization, avoiding unnecessary data long-term occupation of resources; by inputting the remaining storage duration, data type and first storage priority into the trained preset regression model, the predicted access frequency is obtained, which is used to predict the future access demand of construction data; by combining the predicted access frequency of the first construction data and the second storage priority of the process to which the first construction data belongs, the predicted data heat is calculated, which more accurately reflects the actual value of new data in the current construction stage, to ensure that high-frequency construction data of key processes obtain storage resources preferentially; by analyzing the spatio-temporal correlation degree of the second construction data and the process to which the first construction data belongs, the third storage priority of the old data is dynamically calculated, so that the storage value of the old data is adjusted in real time with the advancement of new processes, avoiding the mis-elimination of associated data with reference value to the current construction due to isolated evaluation of old data; by combining the predicted access frequency of the second construction data and the third storage priority to calculate the predicted data heat, the value evaluation of the old data is integrated with the correlation degree of the new process, to ensure that the old data with supporting effect on the current construction obtain reasonable storage priority; by updating the data in the preset storage according to the predicted data heat, high-value data is retained or added to the preset storage, and low-value data is eliminated, realizing the on-demand allocation of storage resources, and improving the utilization rate of storage resources and the future utilization rate of the data stored in the preset storage.

[0099] The above is only a preferred embodiment of the present application, and does not limit the present application in any form. Although the present application has been disclosed as above, it is not intended to limit the present application. Any skilled person in the art can make some changes or modifications to the above disclosed technical content without departing from the scope of the technical solution of the present application, and any simple modification, equivalent change and modification of the above embodiments according to the technical essence of the present application are still within the scope of the technical solution of the present application.

Claims

1. A method for optimizing storage of smart construction data, characterized in that, The intelligent construction data optimization storage method comprises the following steps: S1, upon receiving a data storage request, obtaining a remaining storage duration corresponding to each group of first construction data and each group of second construction data according to the collection time and data type of each group of first construction data to be stored and each group of second construction data already stored in a preset storage; S2, inputting the remaining storage duration, data type and first storage priority corresponding to the data type of each group of first construction data and each group of second construction data into a trained preset regression model respectively, and obtaining a predicted access frequency corresponding to each group of first construction data and each group of second construction data; S3, obtaining a predicted data heat corresponding to each group of first construction data according to the predicted access frequency corresponding to each group of first construction data and the second storage priority corresponding to the construction process to which the first construction data belongs; S4, obtaining a third storage priority corresponding to each group of second construction data according to the construction process to which each group of second construction data belongs and the corresponding second storage priority; S5, obtaining a predicted data heat corresponding to each group of second construction data according to the predicted access frequency corresponding to each group of second construction data and the third storage priority; S6, updating the construction data stored in the preset storage according to the predicted data heat corresponding to all first construction data and second construction data.

2. The intelligent construction data optimization storage method according to claim 1, wherein, S1 comprises the following steps: S11, taking each group of first construction data and each group of second construction data as reference construction data; S12, for any group of reference construction data, obtaining a plurality of groups of historical construction data corresponding to the current data type and the actual storage duration corresponding to each group of historical construction data according to the data type corresponding to the current reference construction data; S13, obtaining a reference storage duration corresponding to the current data type according to the actual storage duration corresponding to each group of historical construction data; S14, obtaining a stored duration corresponding to each group of reference construction data according to the collection time corresponding to each group of reference construction data and the current time; S15, obtaining a remaining storage duration corresponding to each group of reference construction data according to the stored duration corresponding to each group of reference construction data and the corresponding reference storage duration. 3.The intelligent construction data optimization storage method of claim 2, wherein, The intelligent construction data optimization storage method further comprises the following steps: S10, obtaining a plurality of groups of historical construction data and a plurality of accessed times corresponding to each group of historical construction data; S20, for any group of historical construction data, obtaining a plurality of remaining storage durations corresponding to the current historical construction data and the accessed frequency in each remaining storage duration according to the collection time corresponding to the current historical construction data and the plurality of accessed times, and the reference storage duration corresponding to the data type corresponding to the current historical construction data; S30, taking each remaining storage duration corresponding to the current historical construction data, the data type and the first storage priority corresponding to the data type as a group of training samples, and taking the accessed frequency corresponding to each remaining storage duration as the training label of the corresponding training sample; S40, traverse all group historical construction data, obtain all training samples and training labels corresponding to each training sample; S50, train the initial regression model according to all training samples and training labels corresponding to each training sample, and obtain the trained preset regression model. 4.The intelligent construction data optimization storage method of claim 3, wherein, S20 comprises the following steps: S201, according to the acquisition time corresponding to the current historical construction data and the reference storage duration, obtain the reference storage time point corresponding to the current historical construction data; S202, according to the acquisition time corresponding to the current historical construction data and the i th preset step length corresponding to the data type corresponding to the current historical construction data, obtain the i th reference time point corresponding to the current historical construction data, wherein the total number N of reference time points corresponding to the current historical construction data = round down (T / Δt), T is the reference storage duration corresponding to the current historical construction data, Δt is the preset step length corresponding to the data type corresponding to the current historical construction data, round down ( ) is the down rounding function, i = 1, 2, … N; S203, according to the i th reference time point corresponding to the current historical construction data and the reference storage time point, obtain the i th residual storage duration; S204, according to the several accessed time corresponding to the current historical construction data, determine the accessed number corresponding to the current historical construction data in the i th residual storage duration as the accessed number corresponding to the current historical construction data in the i th reference time point to the reference storage time point. 5.The intelligent construction data optimization storage method of claim 1, wherein, S3 comprises the following steps: S31, for any group of first construction data, the product of the predicted access number corresponding to the current first construction data and the second storage priority corresponding to the construction process to which the current first construction data belongs is determined as the predicted data heat corresponding to the current first construction data; S32, traverse all groups of first construction data, obtain the predicted data heat corresponding to each group of first construction data. 6.The intelligent construction data optimization storage method of claim 1, wherein, S4 comprises the following steps: S41, obtain the time parameter and space parameter corresponding to each construction process; S42, for any group of second construction data, according to the time parameter corresponding to the construction process to which the current second construction data belongs, and the time parameter corresponding to the construction process to which each group of first construction data belongs, obtain the time correlation degree between the current second construction data and each group of first construction data; S43, according to the space parameter corresponding to the construction process to which the current second construction data belongs, and the space parameter corresponding to the construction process to which each group of first construction data belongs, obtain the space correlation degree between the current second construction data and each group of first construction data; S44, the time correlation degree and the space correlation degree between the current second construction data and each group of first construction data, and the second storage priority corresponding to each group of first construction data, obtain the third storage priority corresponding to each group of second construction data.

7. The intelligent construction data optimization storage method of claim 1, wherein, S5 comprises the following steps: S51, for any group of second construction data, the product of the predicted access number corresponding to the current second construction data and the third storage priority is determined as the predicted data heat corresponding to the current second construction data; S52, traverse all groups of second construction data to obtain predicted data hotness corresponding to each group of second construction data. 8.The intelligent construction data optimization storage method of claim 1, wherein, S6 includes the following steps: S61, sort all first construction data and second construction data in descending order of predicted data hotness to obtain data sorting results; S62, obtain storable construction data according to the data amount corresponding to each group of first construction data and each group of second construction data, and the storage space corresponding to the preset memory; S63, update the storable construction data to the construction data stored in the preset memory.

Images