Case-based reasoning (CBR), which is based on the cognitive assumption that similar problems have similar solutions, is an important problem-solving and learning method in the field of artificial intelligence (AI). In this article, the development of CBR is reviewed, and the major challenges of CBR are summarized. The paper is organized into four parts. First, the basic framework and concepts of CBR are introduced. Then, the developed technology and innovative work that were designed to solve problems by CBR are summarized. Then, the application fields of CBR are summarized. Finally, according to the idea of deep learning and interpretable AI, the main challenges for the future development of CBR are proposed.

To accurately predict the furnace temperature and flue gas oxygen content of the municipal solid waste incineration (MSWI) process under variable operating conditions, a data-driven multitarget online predictive modeling method based on an improved stochastic configuration network (SCN) is proposed in this article. This method configures the input weights and biases of the new hidden layer nodes of the SCN through a competitive guidance strategy. The model output weights are sparsely constrained using a matrix elastic net, the modeling accuracy is improved by using the correlation between the furnace temperature and flue gas oxygen content. On this basis, the output weights of the model are recursively updated using a sparse constrained direction forgetting algorithm to improve the online prediction accuracy. Finally, the performance of the proposed method is validated using historical data from the MSWI process. The experimental results show that the proposed multitarget prediction model is adaptive and can accurately predict the trends of furnace temperature and flue gas oxygen content under variable operating conditions.

Fault detection in the waste incineration process depends on high-temperature image observation and the experience of field maintenance personnel, which is inefficient and can easily cause misjudgment of the fault. In this paper, a fault detection method is proposed by combining stochastic configuration networks (SCNs) and case-based reasoning (CBR). First, a learning pseudo metric method based on SCNs (SCN-LPM) is proposed by training SCN learning models using a training sample set and defined pseudo-metric criteria. Then, the SCN-LPM method is used for the case retrieval stage in CBR to construct the fault detection model based on SCN-CBR, and the structure, algorithmic implementation, and algorithmic steps are given. Finally, the performance is tested using historical data of the MSW incineration process, and the proposed method is compared with typical classification methods, such as a Back Propagation (BP) neural network, a support vector machine, and so on. The results show that this method can effectively improve the accuracy of fault detection and reduce the time complexity of the task and maintain a certain application value.

To improve the accuracy of similarity measures in case-based reasoning, in this paper, we propose a deep metric learning method based on a self-attention mechanism and a Siamese neural network to realize the weighted similarity measure between cases. In this method, weight assignment and similarity measurement processes are integrated into a deep network. The method can map cases to a new feature space through nonlinear processing of the network layer to obtain better feature representation. The inner relationship between the features is captured by the self-attention mechanism, which is connected to the previous network layer, and the weight of the features is determined by the scoring function. Finally, a metric function is added to the contrastive loss to measure the case similarity. Experiments show that the accuracy of this method is better than that of other algorithms in the similarity measure and can improve the accuracy of case retrieval.

It is difficult to establish an accurate mechanism model for prediction incinerator temperatures due to the comprehensive complexity of the municipal solid waste (MSW) incineration process. In this paper, feature variables of incineration temperature are selected by combining with mutual information (MI), genetic algorithms (GAs) and stochastic configuration networks (SCNs), and the SCN-based incinerator temperature model is obtained simultaneously. Firstly, filter feature selection is realized by calculating the MI value between each feature variable and the incinerator temperature from historical data. Secondly, the fitness function of GAs is defined by the root mean square error of the incinerator temperature obtained by training SCNs, and features obtained by MI methods are searched iteratively to complete the wrapper feature selection, where the SCN-based incinerator temperature prediction model is obtained. Finally, the proposed model is verified by MSW incinerator temperature historical data. The results show that the SCN-based prediction model using the hybrid selection method can better predict the change trend of incinerator temperature, which proves that the SCNs has great development potential in the field of prediction modeling.