A Methodology for Discriminant Time Series Analysis Applied to Microclimate Monitoring of Fresco Paintings

Multivariate Characterization of Temperature Fluctuations in a Historical Building Using Energy-Efficient IoT Wireless Sensors

Adequate thermic conditions are required for the preventive conservation of artworks, but such optimum conditions cannot always be achieved in historical buildings such as ancient churches. In those cases, it is of interest to assess the potential risk of punctual changes in indoor environments that can be harmful to artworks. These conditions can be assessed by means of a microclimate monitoring system comprised of a set of energy-efficient wireless sensors connected to the cloud using IoT techniques. This approach was followed at the baroque church of Saint Thomas and Saint Philip Neri in Valencia (Spain). A set of 26 wireless nodes was installed, which recorded values of temperature and relative humidity every hour for a period of 7 months. Small differences of temperature were obtained among sensors, so that an efficient methodology based on principal component analysis (PCA) was applied for the characterization of similarities and dissimilarities between sensors. Daily ranges of temperatures were studied as well as mean trajectories, differences between days of the week, and changes in the correlation structure of daily median values over time. Results provide a framework for an efficient characterization of temperatures in heritage buildings based on a network of wireless sensors. Such a framework is useful to assess the potential risk of temperature fluctuations on the preventive conservation of historical buildings and artworks.

Characterization of Temperature Gradients According to Height in a Baroque Church by Means of Wireless Sensors

The baroque church of Saint Thomas and Saint Philip Neri (Valencia, Spain), which was built between 1727 and 1736, contains valuable paintings by renowned Spanish artists. Due to the considerable height of the central nave, the church can experience vertical temperature gradients. In order to investigate this issue, temperatures were recorded between August 2017 and February 2018 from a wireless monitoring system composed of 21 sensor nodes, which were located at different heights in the church from 2 to 13 m from the floor level. For characterizing the temperature at high, medium and low altitude heights, a novel methodology is proposed based on sparse Partial Least Squares regression (sPLS), Linear Discriminant Analysis (LDA), and the Holt-Winters method, among others, which were applied to a time series of temperature. This approach is helpful to discriminate temperature profiles according to sensor height. Once the vertical thermal gradients for each month were characterized, it was found that temperature reached the maximum correlation with sensor height in the period between August 10th and September 9th. Furthermore, the most important features from the time series that explain this correlation are the mean temperature and the mean of moving range. In the period mentioned, the vertical thermal gradient was estimated to be about 0.043 ∘C/m, which implies a difference of 0.47 ∘C on average between sensor nodes at 2 m from the floor with respect to the upper ones located at 13 m from the floor level. The gradient was estimated as the slope from a linear regression model using height and hourly mean temperature as the predictor and response, respectively. This gradient is consistent with similar reported studies. The fact that such gradient was only found in one month suggests that the mechanisms of dust deposition on walls involved in vertical thermal gradients are not important in this case regarding the preventive conservation of artworks. Furthermore, the methodology proposed here was useful to discriminate the time series at high, medium and low altitude levels. This approach can be useful when a set of sensors is installed for microclimate monitoring in churches, cathedrals, and other historical buildings, at different levels and positions.

Multivariate Time Series Analysis of Temperatures in the Archaeological Museum of L'Almoina (Valencia, Spain)

An earlier study carried out in 2010 at the archaeological site of L’Almoina (Valencia, Spain) found marked daily fluctuations of temperature, especially in summer. Such pronounced gradient is due to the design of the museum, which includes a skylight as a ceiling, covering part of the remains in the museum. In this study, it was found that the thermal conditions are not homogeneous and vary at different points of the museum and along the year. According to the European Standard EN10829, it is necessary to define a plan for long-term monitoring, elaboration and study of the microclimatic data, in order to preserve the artifacts. With the aforementioned goal of extending the study and offering a tool to monitor the microclimate, a new statistical methodology is proposed. For this propose, during one year (October 2019–October 2020), a set of 27 data-loggers was installed, aimed at recording the temperature inside the museum. By applying principal component analysis and k-means, three different microclimates were established. In order to characterize the differences among the three zones, two statistical techniques were put forward. Firstly, Sparse Partial Least Squares Discriminant Analysis (sPLS-DA) was applied to a set of 671 variables extracted from the time series. The second approach consisted of using a random forest algorithm, based on the same functions and variables employed by the first methodology. Both approaches allowed the identification of the main variables that best explain the differences between zones. According to the results, it is possible to establish a representative subset of sensors recommended for the long-term monitoring of temperatures at the museum. The statistical approach proposed here is very effective for discriminant time series analysis and for explaining the differences in microclimate when a net of sensors is installed in historical buildings or museums.